PROCEEDINGS OF THE UNITED STATES NATIONAL MUSEUM
lylhe
SMITHSONIAN INSTITUTIONU. S. NATIONAL MUSEUM
Vol. 92 Washington: 1942 No. 3153MEXICAN HERPETOLOGICAL MISCELLANY
By HoBART M. Smith
Studies on the reptiles from Mexico in the United States NationalMuseum have brought to light a number of unnamed races and spe-cies. Some of them have been diagnosed elsewhere, and the presentcompilation includes the remaining notes that appear worthy of pre-liminary publication in advance of the contemplated complete sum-mary. A large portion of the material on which the present notesare based was collected and studied during my tenure of the WalterRathbone Bacon Traveling Scholarship of the Smithsonian Insti-tution, to whose authorities I am greatly indebted. ""?;..? yThe following notes have been segregated under nine subtitles >*1.?Six new species and subspecies of Sceloporus, with a redefinition of theformosus group.2.?A new horned lizard from Durango.3.?^A tentative arrangement and key to Mexican Oerrhonotus, with the de-scription of a new race.4.?An unnamed Cclcstus from Mexico, with a key to mainland species ofthe genus.5.?New xantusiid lizards.6.?The Mexican subspecies of Drymobius margaritiferus.7.?Notes on Mexican Imantodes.8.?Two new snakes of the genus Clelia.9.?Additional notes on ConopMs.1.?SIX NEW SPECIES AND SUBSPECIES OF SCELOPORUS, WITH AREDEFINITION OF THE FORMOSUS GROUPMaterial recently obtained in Mexico, particularly by my wife andme in 1939 and 1940, during my tenure of the Walter RathboneBacon Traveling Scholarship of the Smithsonian Institution, has469013?42 1 , 349
350 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92
necessitated a redefinition of the entire formosiis group, as well as ofthe subspecies of jarrovii in the poinsettii group. One other sub-species, anticipated when the review of the Mexican species waswritten,^ has been defined on the basis of material collected by Dr.E. .H, Taylor at Omilteme, Guerrero. Still another race, also an-ticipated previously, has been defined in melanorhinus of the spinosusgroup.I am much indebted to Dr. Taylor for assistance in collectingnumerous specimens, for the gift of several obtained by him inregions not visited by me, and for the loan of his own material. Iam also greatly indebted to Dr. L. C. Stuart for permitting me toexamine material recently collected by him in Guatemala, withoutwhich a satisfactory allocation of northern Central American mem-bers of the formosus group would have been impossible.SCELOPORUS STEJNEGERI, new speciesHolotype.?V.^.'^M. No. 112634, an adult male from Tierra Colo-rada, Guerrero.Paratypes-.-^T^Qwty, including U.S.N.M. Nos. 112635-112648, andEHT-HMS Nos. 22285-22287, 27299-27301, all topotypes, collectedby E. H. Taylor, Kichard Taylor, and H. M. Smith.Diagnosis.?A member of the fonnotiiis group, with a complete orvery nearly complete, black nuchal collar, not light bordered ; supra-oculars large, seldom with an outer row of small scales, all separatedfrom median head scales; frontonasals normal, in contact with eachother; 2 canthals; dorsals 37 to 42; scales around body 45 to 50;femoral pores 16 to 21; generally 4 scales (minimum count) frommedian frontonasal to rostral.Description of holotype.?Head somewhat elongate, not shortenedand thickened as in spinosus group; interparietal relatively large,larger than entire frontal ; parietal short, subtriangular, not reachingposterior edge of interparietal ; a small, rectangular frontoparietal oneach side; posterior section of frontal in contact with interparietal,less than a third size of anterior section of frontal; frontal ridgesprominent; five large supraoculars, separated from median headscales by one row of small scales, from superciliaries by one completerow of small scales and by one or two tiny, extra scales ; small, azygousscale separating prefrontals medially; frontonasals large, in contactwith each other, median the largest; a large (mutilated), trans-versely elongate scale in front of median frontonasal; three scalesfrom latter to rostral; two canthals; a subnasal, loreal, and singlepreocular; subocular elongate, followed by two keeled postoculars;six superciliaries, five visible from above.
? Field Mus. Nat. Hist., zool. ser., vol. 26, 1939,
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 351Snout somewhat mutilated; labiomentals not reaching mental;gular scales notched posteriorly; temporal scales keeled, mucronate;no distinct auricular lobules.Dorsal scales keeled, strongly mucronate, with several lateralmucrones, 39 from occiput to base of tail ; 47 scales around middle ofbody ; femoral pores 18-19 ; 20-21 lamellae under fourth toe.Snout to vent 94 mm. ; tail 128 mm. ; snout to posterior border ofear 21 mm. ; fourth toe 21 mm.Color.?Bluish green above (yellowish where scales are not shed),unmarked save by a broad, black, nuchal collar, complete dorsally,and narrowly continuous around throat; digits and tail not notablybarred. Throat greenish yellow anteriorly, becoming dark blue pos-teriorly ; chest white ; sides of belly dark blue, these areas reaching togroin and nearly to axilla but not onto thighs, and separated mediallyby a narrow light area only two or three scales wide ; blue patches notdark bordered medially ; ventral surfaces of limbs bluish.Vanation.?The 20 paratypes have the frontoparietals separated bycontact of frontal and interparietal, except in one in which they areseparated by a small azygous scale. The prefrontals are in contact in3, separated by contact of frontal and median frontonasal in 3, andseparated by an azygous scale in 14. The canthals are 2 in all, theanterior never above the canthal ridge. The frontonasals are nor-mal except in 1, in which the median is separated from one of thelaterals; the supraoculars are generally 5, sometimes 4 or 6, usuallyin 1 row, sometimes with a few small scales representing an outerrow. The lorilabials are reduced to 1 row below subocular on 1 sidein 4, on both sides in 8; the remainder have 2 complete rows belowthe subocular. The scales from the median frontonasal to rostralare 4 or 5 (4, 11; 5, 9). Other variations in scale counts are givenin Table 1.The young have faint, diagonal light and dark marks on the sidesof the body. Both young and females have narrow, interrupted,transverse dark bands, about seven on the body and one or two onthe neck. The nuchal collar is regularly present and complete in all,although sometimes not very well defined (in discolored specimens).Comparisons.?This species has a higher femoral pore count thanany other of the group. The only ones with which it may be com-pared are m. aca?ithijius, formosus, and asper. The first rarely has16 femoral pores (its maximum count, occurring 3 times in 84 counts),the supraoculars are generally in contact with the median head scales,and the maximum dorsal scale count is 39. In formosus the dorsalscales reach their maximum count at 37 (4 in 52), the internasals arelarger, there is no single large scale preceding the median fronto-nasal, and the coloration is considerably different. In asper the dor-sals are 35 or less, and the coloration is much different.
352 PROCEEDINGS OF THE NATIONAL MUSEUISIHabits.?The species was found only on the boulders m the amaz-ingly rugged canyon at the city limits of the small town of TierraColorada. They are not common, and are exceedingly wary; infact, only one adult male has been obtained.Remarks.?The name stejnegeri appeared as a nomen nudum in1939 ^. It was a lapsus for some other name, the identity of whichis not apparent. It is a pleasure to fix the name for a species sodistinct from others.Table 1.
?
Variation in scale counts of stejnegeri
Catalog number
MEXICAJNT HERPETOLOGICAL MISCELLANY?SMITH 353dorsal scale rows; posterior portion of throat (males) black, scalesin median area black-edged with blue centers, scales anteriorly paleblue; no yellow or orange on throat; females and young of bothsexes with parallel, longitudinal, alternating light and dark lineson neck.Description of holotype.?Dorsal scales 30; scales around body 39;femoral pores 14r-14; 2 canthals.Color.?Dorsal surface of body brilliant green; continuous, parallel,longitudinal black lines following the edges of the dorsal scale rows;dorsal surface of head black, with a light spot on each parietal, onthe interparietal, posterior section of frontal, both prefrontals,lateral frontonasals, posterior pair of internasals, and on several ofthe supraoculars and superciliaries ; a large, black shoulder patch oneach side, the two separated from each other by six scale rows; theblack of shoulder patches continuous around neck; scales on pos-terior part of throat edged with black or very dark blue, the centerslight blue; scales on anterior part of throat and chin pale blue withdarker edges. Chest, a broad line down middle of abdomen, ventralsurfaces of limbs and tail and preanal region all slightly bluish;sides of abdomen dark blue, and these areas with a narrow, poorlydefined, black median border.Variation.?Females lack the brilliant green and blue color of themales, but may have light spots on the head.In the young the back is more or less uniform gray or slate ; on theneck is a median longitudinal light line extending from the upperedge or slightly above the ear to the upper edge of the black shoulderpatches ; these are bordered medially by a narrow dark line of similarextent ; these in turn bordered by a light line, which is separated fromits mate by a median dark line; another light line extends from theposterior portion of the supralabial region through the ear and termi-nates abruptly on the middle of the neck ; below this is another lightline from ear to the black shoulder patches.Females are marked much like the young, except that the neckmarkings are not quite so distinct; most distinct is the dorsolaterallight line and its bordering dark line, from upper edge of ear to upperedge of the shoulder patch. Some adult females have light markson the head, similar to males.Gom/parisons.?There are no well-marked differences between/. formosus and /. scitulus in scutellation. The latter form has alower average dorsal count than the former, but the range of variationof the one form overlaps that of the other too extensively to permitseparation of any population on the basis of this character. Twenty-six of /. scitulus show a range from 30 to 34 (30, 4; 31, 7; 32, 4; 34, 4),
354 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92average 31.9 ; while 49 /. fornvosus show a range of from 32 to 37 (32, 8
;
33, 11 ; 34, 9 ; 35, 11 ; 36, 6 ; 37, 4) , average 34.2.The only difference of recognizable significance between the tworaces is in pattern ; in this there is a very striking divergence betweenthem. Males, females, and yonng of /. formosus lack the longitudinallight and dark lines on the neck, and the scattered light spots on thehead, of /. scitulus; and the adult males of the former are blue, notgreen as in the Guerrero race, lack the longitudinal dark lines on thedorsal surface of the body and have a broad yellow or orange areaon the middle of the throat. In males of /. scitulus yellow or orangeon the throat is completely absent, the whole throat being blue.Remarks.?The race /. scitulus is illustrated in The Mexican andCentral American Lizards of the Genus Scelopottts.'^SCELOPORUS PREZYGUS, new speciesHolotype.?V.S.'N.'M No. 46881, from Conjab, 5,300 feet, Chiapas(between San Bartolome and Comitan) . Collected by E. W. Nelsonand E. A. Goldman.Diagnosis.?A member of the formiosus group, with 31 dorsal scales,supraoculars in 2 rows, 13 to 14 femoral pores, 2 canthals and medianfrontonasals separated from lateral frontonasals.Desci'iption of holotype.?Head somewhat flattened and elongate;interparietal about three times as large as a parietal ; a single parietalon each side, but little larger than a frontoparietal; latter single oneach side ; frontal narrowly in contact with interparietal ; supraocularsin two rows ; the scales of the inner row about twice as large as thoseof outer row, which is composed of three scales ; supraoculars separatedfrom median head scales by a complete row of granular scales, fromsuperciliaries by one complete and another incomplete row of smallscales; frontal in contact with median frontonasal; latter consider-ably larger than lateral frontonasals, separated from them by a smallscale ; scales in internasal area large, keeled, pitted, three from medianfrontonasal to rostral ; two canthals ; a large, elongate subnasal ; threesmall loreals; preocular longitudinally divided, a large upper andsmall lower scale : an elongate subocular and two keeled postocularstwo complete rows of lorilabials below subocular; four supralabialsand five infralabials to a point below middle of eye.Two rows of labiomentals, the outer not reaching mental, the innerextending anteriorly to a point even with the suture between the sec-ond and third infralabials ; posterior gulars notched ; temporal scaleskeeled, very feebly mucronate; three small auricular lobules, upperlargest; scales between ear and lateral nuchal pocket more strongly
* Field Mu3. Nat. Hist., zool. ser., vol. 26, pi. 1, 1939.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 355keeled and mucronate, with a number of fine, lateral mucrones ; scalesbetween lateral nuchal pocket and foreleg keeled, rather stronglymucronate, the largest larger than scales posterior to ear or in tem-poral region.Dorsal scales keeled, mucronate and denticulate to a moderate de-gree, 31 from occiput to base of tail ; 36 scales around middle of body;13 to 14 femoral pores; 22 lamellae under fourth toe; snout tovent 77 mm. ; tail 112 mm. ; snout to posterior margin of ear 19.9 mm.
;
snout to occiput 15.9 mm.; fourth toe 21.7 mm.; hind leg 55.5 mm.Color.?Dorsal surface yellowish gray (slightly bluish where scalesare shed) ; a narrow, black nuchal collar, complete on dorsal surface,scarcely visible on throat; collar not extending onto arm, not overtwo scales in width. Dorsal surface otherwise unmarked. Gularregion and chin uniform blue, except a gray area about throat infront of chest; latter white; sides of abdomen apparently lavender,with a broad border covering a width of three scales; latter extend-ing from groin nearly to a line even with axilla, but not extendinglaterally anteriorly to enter axilla. Ventral surfaces of limbs and tailwhite.Remarks.?This specimen was referred by me in 1936 ^ to serrifer.,although several of its peculiarities were pointed out; and in 1939^it was referred to mu.cronatus omiltemanus. Recent material fromChiapas and Guatemala has shown that the characters previouslyconsidered as anomalous have considerable significance. This speci-men does not belong to serrifer, although the belly pattern is thesame, since it has the supraoculars in two rows and the frontonasalsseparated from each other. The latter species is the nearest one of thepoinsettii group ; the more remote omiltemanus and its relatives, al-though having similar supraoculars, have the frontonasals in contactand a different belly pattern. In fact, it is believed that prezygusbelongs to the fonnosiis rather than to the poinsettii group, since ithas a very narrow collar not light bordered (apparently). It differswidely from others of the formosus group, however, in having largedorsals, the frontoparietals separated from each other, supraocularsin two rows, etc. It appears to be a link between the formosus andpoinsettii groups, in the same manner as lunaei and acanthinus areobviously links between the formosxis and spinosus groups. All mem-bers of the latter group can be traced to acanthinics and lunaei, whileall of the poinsettii group can be traced to prezygus and serrifer. Thename prezygus refers to the phylogenetic position of this species, nearthe ancestral type of the collared {poinsettii) group of the genus.
" Kansas Univ. Sci. Bull., vol. 25, pp. 561, 562, 1936.'Field Mus. Nat. Hist, zool. ser., vol. 26, p. 221, 1939,
356 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92KEY TO MEMBERS OF THE FORMOSUS GROUP
1. Typically one canthal 2Typically two canthals 42. Supraoculars very large, in a single row, frequently one or more in contactwith median head scales ; dorsals 31 to 38 ; a lowland species ; Motagua Rivervalley and arid basin at Salama, Guatemala lunaeiSupraoculars smaller, in 2 rows, rarely any in contact with median headscales; dorsals 30 to 51; highland races 33. Dorsals 30 to 39 ; El Salvador and central Honduras to Panamamalachiticus malachiticusDorsals 39 to 51 ; central plateau of Guatemala and its extensionsm. smaragdinus4. Median separated from lateral frontonasals 5Median in contact with lateral frontonasals 75. Dorsals about 31 ; internasals large, keeled, 3 from rostral to median fronto-nasal ; known only from Conjab, between San Bartolom6 and Comitdn,Chiapas prezygusDorsals 37 to 45 ; internasals smaller, not keeled, rugose or not 66. A complete nuchal collar, sometimes narrowly interrupted medially ; centralVeracruz to Isthmus of Tehuantepee malachiticus salviniiNuchal collar incomplete; highlands of Chiapas and northern Guatemala inCohan area m. taeniocnemis7. Anterior section of frontal usually longitudinally divided; canthals sharplyridged; no nuchal collar; ventrals a fourth, laterals not more than two-thirds, size of dorsals; mountainous regions of western Mexico from Nayaritto Guerrero asperAnterior section of frontal rarely longitudinally divided ; canthals rounded
;
a nuchal collar or not ; ventrals a half, laterals three-fourths size ofdorsals 88. A broad, nearly or quite complete, nuchal collar 9Collar, if present, restricted to sides of neck 109. Femoral pores 12 to 16 ; one or more of supraoculars generally in contact withmedian head scales; Pacific coast foothills, Chiapas to El Salvadormalachiticus acanthinusFemoral pores 16 to 21; supraoculars not in contact with median head scales;central Guerrero (Tierra Colorada) stejneg'eri10. Males without yellow on throat, dorsal scale rows black-edged, dorsal headscales with a light blue center; central Guerrero formosus scitulusMales with yellow or orange on throat, dorsal surface uniform blue, headscales not light-spotted ; central Veracruz south in mountainous regions tothe Isthmus of Tehuantepee f. formosusSCELOPORUS MUCRONATUS AUREOLUS, new subspeciesEolotype.?^U.S.N.M. No. 112232, male, 2 miles west of Acultzingo,Veracruz.Paratypes.?Twenty-six topotypes (Nos. 112233-112258) in theUnited States National Museum; 36 topotypes (Nos. 3073-3080, 3082-3102, 3171-3174, 3193-3195) in the EHT-HMS collection; 16 speci-mens (No, 1516) from Atzitzintla, Puebla, in the Field Museum ofNatural History; and 6 specimens (Nos. 18815-18820) from the vicin-
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 357ity of Puebla, Puebla, in the American Museum of Natural History.Diagnosis.?Similar to m. ynucronatus, except : Dorsal scales usuallymore than 32 ; femoral pores usually over 13 ; no parallel longitudinaldark lines on middle of throat; adult males usually with longitudinaldark lines following the edges of the dorsal scale rows; centers ofscales light; collar involving not over 4 scale lengths longitudinally.Remarks.?I described this subspecies in detail ^ as mucronatusomiltenmnus. The latter name, however, must be restricted to thelarger scaled, broad-collared form lacking longitudinal dark linesfollowing the dorsal scale rows, that occurs in the isolated mountainrange of central Guerrero.Com{parisons.?S. m. aureolus is amply well differentiated from bothmucronatus mucronatus and m. omiltemanus by having more nu-merous dorsals (30 to 38, average 34.3, 91.6 percent over 31 in 59aureolus; 27 to 32, average 29,6, 93.6 percent 31 or less in 49 mucro-natus-^ 29 to 32, average 30, 92.3 percent less than 32 in omiltemanus) ^and a lined dorsal pattern in adult males. From m. mucronatus italso differs in average femoral pore count (11 to l7, average 14, in124 counts of aureolus ; 10 to 17, average 12.8, in 94 counts of mucro-natus'^ 11 to 16, average 13.8, in 26 counts of omiltemanus) and inlacking the parallel, longitudinal dark lines on the throat (presentin all except adult male mucronatus, lacking in omiltemanus) . It isapparent that m. mucronatus and m. omiltemanus are more like eachother than either is like aureolus. They differ from each other indorsal pattern ; omiltemanus lacks the large dorsal blotches and theparallel, longitudinal dark lines on the throat of mucronatus, and inaddition has a collar involving at least 4i/^ scale lengths (4 or lessin mucronatus).SCELOPORUS JARROVII SUGILLATUS, new subspeciesHolotype.?U.S.N.M. No. 112100, male, from the edge of the eastend of Lake No. 4, Zempoala, Mexico, Mexico.*Paratypes.?Fifty-four, all topotypes, including U.S.N.M. Nos.112072-112099, 112101-112111, and EHT-HMS Nos, 22311-22321.Diagnosis.?Supraoculars in 2 rows ; dorsal scales 37 to 44, average40.6; median lateral body scales distinctly smaller than middorsalscales ; scales on^ dorsal surface of upper foreleg about twice as largeas those on lower foreleg; gray or blue-gray above, streaked, with arather broad, black nuchal collar (6 to 8 scales wide) ; sides of bellyslate blue, not black-edged, with vertical streaks of black; chin lightblue ; males and females practically indistinguishable in ventral color.
' Kansas I'niv. Sci. Bull., vol. 24, pp. 59]-598, text fig. 12, pi. 50, fig. 1, 1038.The boundary line between the states of Morelos and Mexico passes through the NationalPark of Zempoala. Most of the park is in the State of Mexico.
358 PROCEEDll^GS OF THE NATIONAL MUSEUM vol. 92Description of holotype.?Interparietal a little less than three timessize of parietal ; one frontoparietal on each side, separated from eachother by a small median scale; posterior section of frontal a littlemore than half size of anterior ; two prefrontals in contact medially
;
three frontonasals, the median not separated from the others; inter-nasals irregular; four postrostrals ; two canthals; a subnasal and anelongate loreal ; preocular divided transversely ; an elongate subocularand two postoculars; two rows of lorilabials, complete below eye;two rows of supraoculars, the scales in the outer row about one-halfto one-third size of scales in inner row ; outer row of labiomentals notreaching mental; inner row reaching to suture between second andthird infralabials; four supralabials to a point below middle of eye;most of gular scales with a single apical notch; those near gularfold area with two notches.Temporal scales keeled, rather strongly mucronate; six auricularlobules on one side, four on other; scales between ear and lateralnuchal pocket about half size of dorsal scales, larger than scalesbetween lateral nuchal pocket and arm, keeled and stronglj' mucronate,weakly denticulate. Dorsals weakly keeled, mucronate, feebly den-ticulate (more strongly anteriorly) ; lateral scales a little more stronglymucronate and more denticulate than dorsal scales; scales distinctlydecreasing in size laterally, those halfway between axilla and groinsmaller than middorsal scales ; 41 scales from occiput to base of tail46 scales around middle of body.Scales on dorsal surface of upper foreleg nearly twice as large asthose on lower foreleg, all keeled, mucronate and denticulate; dorsalscales on shank and thigh subequal to each other and to scales ondorsal surface of upper foreleg, a little smaller than dorsal scaleson body, and a little more strongly keeled and mucronate; dorsalscales on tail considerably larger than dorsals on body, more stronglykeeled and mucronate; subcaudals smooth except toward tip of tail;femoral pores 15-16.Snout to vent 85 mm. ; tail 125 mm. ; tibia 15 mm. ; snout to occiput16 mm. ; snout to posterior border of ear 20 mm.Color.?Adult male : Entire dorsal surface dark slate gray, withoutmarkings save a broad, black nuchal collar covering eight scalesmedially; collar bordered on either edge by a narrow, light, un-interrupted band one scale wide ; digits feebly banded ; tail with veryindistinct dark bands distally. Ventral surface of head dark blue;black nuchal collar extending onto gular region, almost interruptedmidventrally ; area between arm insertions and a broad median lineon belly mostly light ; sides of belly, from axilla to groin, blue-gray,not black-bordered medially; numerous, distinct, transverse blackstreaks on sides of belly, not extending as far medially as the bluish
MEXICAN HERPETOLOGICAL MISCELLANY SMITH 359patches ; groin nearly black ; ventral surfaces of limbs and tail light.Variation.?In 44 specimens the dorsals vary from 37 to 44, average40.3 ; in 64 counts the femoral pores vary from 14 to 18, average 16. Inall the dorsal scales on the lower foreleg are constantly about half thesize of those on the upper foreleg.Adult females are so remarkably like the males in ventral as wellas dorsal coloration that they can be distinguished only with difficultyby this means. The largest measures 77 mm. snout to vent ; the largestmale measures 87 mm.Comparisons.?This is one of the most distinct of the races ofjarrovii, having peculiarities both in pattern and scutellation. Thetransverse black streaks on the sides of the belly do not occur in anyother form of jarrovii/ in fact the only other species having similarmarks is dugesil, a species very different from jarrov'd in the conforma-tion of the scales (mucrones arising within the edges of the scales onsides of body and neck). In no other subspecies of jarrovii do thefemales have the same ventral coloration as the males; such a phe-nomenon is of rare occurrence in Sceloporus., and has been observed inno other species of the poinsettii or related groups.The most obvious peculiarity in scutellation is the relatively smallsize of the dorsal scales on the lo^^er foreleg, compared with those onthe upper foreleg (nearly twice as large). This character will sepa-rate it from all other subspecies of jarrovii. It agrees with iniTWU-eronatu^ in having the lateral scales distinctly decreasing in size, anddistinctly smaller than the dorsals. In the other subspecies the lateralscales, at a point about halfway between axilla and groin, are sub-equal to or larger than the dorsals ; they do not distinctly increase insize m.edial to this point.Habits.?These specimens were found sunning themselves on therocks scattered on the eastern edge of Lake No. 4, at the Lakes of Zem-poala, in the State of Mexico (reached via Tres Cumbres, Morelos).The elevation is about 10,000 feet above sea level. The lizards are ex-tremely wary, so that it was necessary to stalk them from behindboulders in order to get within gunshot of them. Although the rocksand cliffs around several other lakes on the park were examined,sugillatus was not found, although /. ferrariperezi was very abundant.The latter was excluded in the spots where the former was found.KEY TO THE SUBSPECIES OF SCELOPORUS JARROVII1. Lateral body scales distinctly decreasing in size laterally, at a point halfwaybetween axilla and groin, distinctly smaller than dorsal scales 2Lateral body scales not decreasing in size laterally at least up to a pointhalfway between axilla and groin, where they are still subequal to, or evena little larger than, middorsal scales 32. Dorsal scales on lower foreleg about half size of those on upper foreleg ; darktransverse streaks in lateral belly patches ; nuchal collar covering six scales
360 PROCEEDINGS OF THE NATIONAL MUSEUM vor.. 92
medially, or more; high mountains near the southern edge of the plateau,known only from Zempoala, Morelos and Mexico jarrovii sugillatusDorsal scales on lower foreleg but little if any smaller than those on upperforeleg; no dark streaks in lateral belly patches; nuchal collar less thanfour scales long middorsally ; eastern mountains of Mexico, from northernQuer^taro south into Hidalgo and possibly central Veracruzjarrovii immucronatus3. Supraoculars essentially in one row ; if an outer row is evident, it is composedof scales much smaller than those of inner row, and usually number nomore than two; mountains of western Mexico, from Nayarit north intosouthern Arizona and New Mexico jarrovii jarroviiSupraoculars in two rows, those of outer i-ow a little smaller than those ofinner, usually numbering three or more 44. Adult males black above and below, with orange areas and spots on sides ofhead, belly, and tail; only throat, underside of tail, and posterior surfaceof hind leg not black ; females somewhat similar, very dark above, the collarpoorly defined; young with poorly defined, narrow, light borders on neckcollar ; dorsal scales average 37.5 ; southeastern Coahuila jarrovii oberonAdult males light brown above, with very broad, very well defined nuchalcollar ; a median area on belly white except in very largest males ; sidesof abdomen blue, black edged ; females and young with more distinct lightborders on nuchal collar; dorsal scales average 40.6; central plateau regionfrom central Mexico (State) north to northei'n Zacatecas? jarrovii minorSCELOPORUS MELANORHINUS CALLIGASTER, new subspeciesHolotype.?V.^.'^M. No. 112201, Acapulco, Guerrero.Paratypes.?\].^.'^M. Nos. 112199-112200, 47732, topotypes; No.112186, Coyiica, Guerrero; Nos. 112202-112203, 4 kilometers north ofApatzingan, Miclioacan ; San Bias, Nayarit (Nos. 51384-51389, 64667)
;
"Guadalajara," Jalisco (Nos. 2492;5-24926) ; CoHma (Nos. 31496,58159); Tamarindo, Guerrero (No. 47731). Also EHT-HMS Nos.8279-8299, Hda. El Sabino, Miclioacan; Nos. 8302-8310; Acapulco;Nos. 8267-8270, Hda. Queseria, Colima; Nos. 8273A, 8271-8278, Hda.Paso del Kio, Colima. Univ. Mich. Mus. Zool. No. 80070, Hda. Paso delEio, Colima ; No. 80069, Hda. Gloria, Colima.Diagnosis.?Similar to 8. TneJuiwrhinus melanorliimis^ except femo-ral pores usually (84 percent) less than 20 on each side; lateral bellypatches usually confluent medially in adult males.Description of holotype.?Adult male, 85 mm, snout to vent ; dorsalscales 27; femoral pores 17-18; lateral belly patches partially confluentmedially.Comparisons.?This subspecies is distinguished from typical m^M-norhinus largely upon the basis of average femoral pore count. Ninety-three counts of 46 specimens from Guerrero north show a range of va-riation from 17 to 24, average 19.2, with 16.1 percent of the counts over20. Fifty-eight counts of 29 specimens of typical melanorhinus fromOaxaca show a range of variation from 18 to 27, average 21.6, with 77.6percent of the counts over 20.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 361All approximately similar comparison can be made of the total porecounts of specimens of the 2 subspecies. In 7n. calligctster the range isfrom 34 to 46, average 38.4, and 19.6 percent of the counts (46) are over40. In m. melanorhinus the range is from 37 to 53, average 42.8, and82.8 percent of the counts (29) are over 40.All counts available for the species were used in calculating percent-ages. The contrasts between the 2 races would be somewhat greater ifspecimens from certain central Guerrero localities (Mexcala, TierraColorada) were omitted as intergrades. The series available are notsufficiently large to demonstrate whether these localities actually areoccupied by intergrading populations, and for this reason the question-able specimens were not omitted. It is noteworthy, however, that theoccurrence of counts over 20 is greater there than elsewhere in m. calli-gaster. Subspecies of other species apparently intergrade also in thisarea: e. g., Uta h. hicarinata and h. anonymovpha; SceJopoms h. hor-ridiLS and h. oligoponis.
362 PROCEEDINGS OF THE NATIONAL MUSEUM vot.. 92PHRYNOSOMA DOUGLASSII BRACHYCERCUM, new subspeciesHolotype.?V.S.^M. No. 23993, from "Diirango," Mexico, collectedby Dr. Edward Palmer, September 1896.Paratypes.?Three topotypes, Nos. 23994r-23996, same collector anddate.Diagnosis.?Similar to Phrynosoma douglassii, but with a tailshorter than head is wide, or, in adults, very slightly longer; gularscales in straight rows slightly diverging posteriorly, the scales smalland tubercular (strongly convex), not fiat; chest scales keeled; hornscf head ver}^ short, postorbitals, temporals, and occipitals subequal insize.DescHption of holotype.?Head a broad, short, conventionalizedheart-shape in dorsal profile ; in lateral profile, postorbital spines high-est, supraocular and internasal areas forming an obtuse angle witheach other (not a curve) ; width of head (26.2 mm.) at widest point intemporal region much greater than length of head from snout to occi-put (15.6 mm.) or to posterior tip of temporal spines (21 mm., in a pro-jected straight line) ; supraocular region flat, with no enlarged scales;posterior border of supraocular region strongly indented medially,marked by a series of enlarged, slightly protuberant scales; each seriesbegins at postorbital spine and extends anteromedially, but fails toreach its mate by 1 scale ; postorbital, occipital, and temporal spinessubequal in size ; 5 scales between occipital spines ; 2 small, flat spines,separated from each other by a scale, preceding occiput; 3 subequaltemporal spines, the posterior slightly separate from others, the seriescontinuing anteriorly as a row of enlarged scales, disappearing belowabout middle of eye.Three posterior infralabials considerably enlarged; all labialskeeled; a series of enlarged, keeled scales in contact with a fewinfralabials antcriorh^ but separated throughout the remainder ofits length by one to three rows of small scales; gulars in very defi-nite, straight rows slightly diverging posteriorly; these scales notflat, but small and convex (tubercular) ; in extreme posterolateralportion of throat the scales are strongly protuberant and conical,and have no free lateral edges.An irregular series of preauricular spines; tympanum exposed; alarge skin fold anterior to lateral nuchal pocket, and a small oneposteriorly, the former surmounted by two series of spines, the latterby one or two spines; a small, vertical series of spines in front ofand above arm insertion.All dorsals keeled, imbricate (except enlarged spines), not granu-lar; among these are scattered, enlarged, keeled spines of varyingsize, diminishing laterally; a single row of lateral spines (fringe) ;in the posterior part of fringe the spines are separated from each
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 363
other, while anteriorly they are in contact; spines on tail few, small;tail very broad at base, but remainder very slender, not taperinginto base; eight longitudinal series of spines at base of tail.Scales on chest rather distinctly keeled; remainder of ventralscales smooth ; femoral pores 16-18 ; 14-16 lamellae under fourth toe.Total length 80 mm. ; tail 24 mm.Head slightly reddish; body dirty straw yellow, with transverse,median, dusky (gray) crossbands; limbs feebly barred; throat, chest,and sides of abdomen with small, round, scattered dark spots.Variation.?The three paratypes are much as the type. All haveconvex gulars, although some of the scales show edges. The keelson the chest scales are very evident even in the smallest specimen.Measurements (in millimeters) of d. brachycercum
Catalog number
364 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92prefrontal, the presence or absence of whicli lias been used as ageneric or group character. In the most primitive ^roup (deppii)of the genus, it varies intraspecifically to a consideralle degree; inspecimens from a single locality of at least two species it is com-pletely absent in some, well formed in others. In the next mostprimitive group {antauges) it varies greatly interspecifically ; threespecies regularly lack it, while the remainder regularly have it.Only in the other three, more highly modified groups does the azygousprefrontal serve as a group character; in the liocephalus andcaeruleus groups it is regularly present, and in the hnbricatus groupit is regularly absent."All five groups of the genus are represented in IVlexico by atleast one species. The deppii group is composed of six in Mexico:deppii^ flnibriatus, gramineus, oaxacae, taeniatus, and ochoterenai.Only four other names have been proposed in this group; diguetifrom Puebla (a synonym of taeniatus^ not of deppii^ of which it wasdescribed as a subspecies), am^tus from Verapaz, Guatemala (ap-parently distinct; the type has an azygous prefrontal, contrary toBoulenger's statement), vasconcelosii from Argueta, Pacific slope ofGuatemala, and 7-honibifer from Panama. The latter is placed in thisgroup with some hesitation, by description alone; I have seen nospecimens. The group is the only one aiboreal ni habits. The flathead, poorly developed lateral fold, elongate dorsal scales, and largesize of the granular scales on the limbs as well as on the sides of thehead and neck, characterize this group. Several species are green.The other 4 groups are composed of terrestrial species. The mostprimitive group {antauges) is composed of apparently 6 species,in Mexico: antauges^ hocourti., gadovH, moderns (type locality ap-parently Orizaba, not Guatemala as guessed by Cope), obscurus^ andvimdiflavus. All but one of these (gadovii) is known from so fewspecimens that further material may show that Fome of these namesare based on variants. All names proposed for Mexican specimensappear to be valid. Extralimital are alfaroi (Costa Rica), iiionticola(Costa Rica), moreZei^^w (Guatemala) and salvadoren.sis (Salvador);the first 2 appear to be identical " and the last 2 also may proveconspecific with each other.^- The group is characterized by havinga moderately well developed lateral fold (not a deep one as in the3 more recent groups) ; essentially a single loreal (sometimes a verysmall accessory loreal) ; supranasals present (sometimes enlarged toappear like a second pair of internasals) ; head scales flat ; super-ciliary series complete; dorsal scales 45 or more.
" One specimen in 83 of the whole group has a finy azygous prefrontal." Wettstein, Sitz. Akad. Wiss. Wien, niath.-uaturw. Kl., Abt. 1, vol. 143, pp. 28-29,figs. 1-9, 1934."Dunn and Emlen, Proc. Acad. Nat. Sci. Philadelphia, vol. 84, pp. 28-30, 1932.
MEXICAN HERPETOLOGICAL MISCELLANY?SIVnTH 365The liocephalus group is composed of u single widely distributedspecies, "which appears to have recognizable races. The names in-femails^ lemniscatus, liocephalus, ophiur'v.s, tessellatus, and ventralishave been proposed and are available in this group. I can distin-guish three races: I. liocephalus, I. ophiurus, and I. infernalis. Theelongate body and tail, and the presence of at least tv.'o large loreals,one following the other, characterize the group; the superciliary se-ries is complete, and the azygous prefrontal is large and regularlypresent.The caeruleus group, with 11 recognized species and subspecies^*is the largest of the genus. The 19 names proposed in the group are
:
burnettii, caeruleus, cedrosensis, forniosa, grandis, ignavus, kingii,marginata, Tnulticarinata, muUifasciata, nanus, nohilis, palmeri, pauci-carinatus, principis, scincicauda, shastensis, webhii, and wiegmanni.Only kingii occurs in mainland Mexico. The group is characterized bythe absence of the supranasal (fused with first pair of internasals, aunique character in Gerrhonotus) , regular contact of rostral andnasal, and regular presence of an azygous prefrontal.The most highly modified group {imhricatus) is confined to Mexico,so far as known at present. It is the only one that has completely lostthe azygous prefrontal. Other characters are: Convex head scales(except plajmfwns) ; reduction of superciliary series, which rarelyreaches beyond the median outer supraocular, leaving the posteriorouter and posterior inner in contact with orbit. Seven names havebeen proposed in this group, and one more is added below ; of theseeight, six can be associated with apparently recognizable species andsubspecies : i. imbricatus, i. adspersus, I. levicolUs, I. ciliaris, planifronSyand rudicollis. Of these the most doubtful are i. adspersus and plani-frons. G. Uchenigeras and oUi-aceus are synonyms of i. imhricatus.GERRHONOTUS LEVICOLLIS CILIARIS, new subspeciesEolotype.?U.S.N.M. No. 47496, from Sierra Guadelupe, Coahuila,collected by E. W. Nelson and E. A. Goldman, April 28, 1898.Parafypes.?Thirteen. Two topotypes, Nos. 47497-8 ; Sierra Madre,Zacatecas (Nos. 46723-46724) ; Inde, Durango (No. 46843) ; mountainfcnear San Luis Potosi, S.L.P. (Nos. 47207-47210) ; and San Felipe,Guanajuato (EHT-HMS Nos. 10418-10421).Diagnosis.?Like levicolUs, with 16 dorsal scale rows, no azygousprefrontal and convex head shields, except : Dorsal scales 40 to 50.average 42.7 ; 2 superimposed loreals or, if 1, obviously the lower fusedwith upper ; loreal in contact with prefrontal ; usually 3 superciliaries,and always at least an anterior superciliary between preocular andanterior supraocular.
" Fitch, American Midi. Nat., vol. 20, No. 2, pp. 381-424, 1938.469013?42 3
36G PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92Description of hoJotype.?Head not flattened, nearly as thick asbroad; an elongate, narrow interparietal in contact with frontal; oneach side a parietal, no larger than scale following; 2 frontoparietalson each side, a little smaller than parietals, the anterior broadly incontact with frontal ; a pair of large prefrontals, and 2 pairs of inter-nasals; 5 large inner supraoculars, the anterior separated from pos-terior internasals by contact of loreal and prefrontal; 3 small outersupraoculars, the posterior bordering orbit immediately in front ofposterior supraocular of inner series, the other 2 not reaching orbit;posterior superciliaries missing, the series represented by 3 scales sepa-rating the anterior 2 outer supraoculars from orbit, and separatingpreocular from anterior inner supraocular; a very elongate supra-nasal ; a single postnasal ; a small lower loreal, bordered above by an-other loreal more than twice as large; a large preocular; two sub-oculars, the posterior elongate ; 3 postoculars ; supralabials 10-10, thelast elongate, none posterior to eye higlier than those anterior to eye
;
3 anterior temporals.Ten infralabials to a point even with posterior supralabial ; aboutfive pairs of chinshields, the scales of the anterior two pairs in contactmedially, the remainder widely separated ; a row of large labiomentals,larger than the infralabials they border, separating all of chinshields,except the anterior and a narrow portion of the second, from infrala-bials.Lateral fold deep, both on neck and on body ; fold on neck enclos-ing a small, nearly scaleless pocket medially; 16 longitudiiial rows ofdorsals ; 12 rows of ventrals ; 42 dorsals from interparietal to posteriormargin of thighs; 6 median rows of dorsals obtusely keeled, otherssmooth.Snout to vent, 116 mm. ; tail regenerated ; snout to posterior borderof ear 29 mm. ; width of head 20 mm. ; depth of head 17 mm. ; hindleg 32 mm. ; foreleg 26 mm.No dark markings above or below ; head reddish yellow ; body brown-ish yellow ; venter cream. No light flecks present.Variation.?All specimens have 16 rows of dorsal scales, and all but1 (with 14) have 12 rows of ventral scales; dorsals from occiput tobase of tail 40 to 50, average 42.7 (40, 1; 41, 3; 42, 5; 43, 3; 46, 1;50, 1). The number of keeled rows of dorsals varies from 6 to 8, andusually is 6. Two have 2-2 loreals, one has 1-2, one has 2-?, and theremainder have 1-1 ; in those specimens with single loreals, the scaleis narrowed and otherwise so shaped that it evidently is formed by thefusion of two scales. The loreal is in contact with the prefrontal inall. Supralabials 9-10 in 1, 10-10 in 7, 10-11 in 2, 11-11 in 3. Theanterior superciliary is present in all, and in all but 1 this scale com-pletely separates the preocular from the anterior supraocular. In
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 3671 specimen there are 4-6 supercilaries ; the series is practically com-plete, and separates all the outer supraoculars, and all but the posteriorinner supraocular, from the orbit. The others have 2 to 4 supercili-aries, except in 1, on 1 side of which only the anterior superciliaryis present (median fused with an outer supraocular on 1 side). Onespecimen has a very small azygous prefrontal.Color.?Some specimens have faint, narrow, darker brown cross-bands, margined irregularly with flecks of white. Others have flecksof white scattered irregularly over the back and head. Otherwise thecoloration is a more or less uniform, brownish yellow.Comparisons.?This race is very similar to levicoUis, having 16dorsal scale rows. It differs from that in having at least an anteriorsuperciliary {levicollis never has more than the median superciliary),which rarely does not separate the preocular from the anterior supra-ocular (preocular always contacts supraocular in levicollis) ; 2 loreals,or a single loreal formed of 2 fused scales (an obviouslj^ single, moreor less rounded loreal in levicollis) ; loreal always in contact withprefrontal (loreal geuerally separated from prefrontal by contact ofsupraocular and posterior internasal) ; and generally fewer dorsals,varying from 40 to 50, average 42.7 (47 to 50, average 49, in levicollis).There is a possibility that this is identical with a specimen describedby Bocourt as planifrons, from "Oaxaca." If so the type probablybears incorrect locality data, as it now seems very doubtful that thespecies that has been known by the name of levicollis (includingmlia7'is) occurs on the Oaxaca highlands. G. planifrons moreover issaid to have perfectly flat head scales, unlike all others of the hnbrica-tus group. It may possibly be an aberrant specimen of some otherspecies, such as ivibrico.tus, the only other member of the group defi-nitely recorded from the Oaxaca region. That no further specimens'like the type have been collected has no bearing on the validity of thespecies, since the very distinct oaxacae also has remained a desideratumat least in the collections examined by me. In view of this doubtconcerning planifrons, I believe it best to retain that name for aspecies distinct from levicollis; to synonymize the two would requirea shift of names (since planifrons was the earlier named) not at allwarranted by present knowledge.KEY TO MEXICAN GERRHONOTUS
1. Arboreal species with flattened heads ; dorsals with feeble or no keels, longertlian broad ; lateral fold poorly defined ; scales on posterior surface ofthigh and on sides of neck large granules ; adults usually greenish ; centralMexico to Panama deppii group 2Terrestrial species seldom found in trees, with thick heads not flattenedalthough frequently widened posteriorly ; temporal and supra-auricularscales never protuberant ; lateral fold moderately to very well deflned
;
368 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92
scales on posterior surface of thigh and on sides of neclc minute or not;dorsals as long as broad or broader, with well-defined keels; adults gen-erally brown or gray 72. A series of projecting scales above ear 3No projecting scales above ear 4S. Scales above ear short, obtusely conical ; an azygous prefrontal ; centralChiapas (near Comitan) ochoterenaiScales above ear very elongate; no azygous prefrontal; all scales betweenfrontal and rostral paired; central Chiapas, Guatemala fimbriatus4. Suboculars missing or reduced to minute scales; only one anterior temporalbordering orbit ; central Guerrero deppiiSuboculars present ; two anterior temporals bordering orbit 55. Area of granules on sides of neck very narrow ; no granular zone in lateralfold oaxacaeArea of granules covering entire sides of neck ; a granular zone in lateralfold? : 66. Dorsals lai'ger, 25 to 29 from interparietal to posterior margin of thighs;scales across nape usually reduced to 4 or 5 in at least 1 transverse row
;
central Veracruz and adjacent Puebla gramineusDorsals smaller, 32 to 34 from interparietal to posterior margin of thighs:scales across nape not reduced to less than 6 in any transverse row ; south-ern Hidalgo, central and northern Puebla taeniatus7. Supranasal plates absent or fused with first pair of internasal scales ; nasalin contract with rostral ; a large azygous prefrontal ; western Chihuahua,probably eastern Sonora {caerulens group) kingiiSupranasal plates present, small or, if enlarged, forming the second pairof scales behind rostral ; nasal rarely in contact with rostral ; a largeazygous prefrontal or not 88. Posterior inner (usually outer also) supraocular in contact with orbit; super-ciliary series short, only anterior portion present (rarely posterior portionpresent, separating two posterior supraoculars from orbit) ; rarely anazygous prefrontal, minute when present; dorsals from interparietal toposterior margin of thighs 27 to 50 imbricatus group 9Posterior supraoculars separated from orbit by a complete series of super-ciliaries; an azygous prefrontal or not; dorsals not less than 45 149. Longitudinal rows of dorsals 14 or less : 10Longitudinal rows of dorsals 16 1210. Dorsal scales less than 35 (27 to 29) ; western Mexico (State) rudicollisDorsal scales 35 to 43 1111. Rows of dorsals 12; western Mexico (State), perhaps adjacent MichoacSnand Guanajuato imbricatus adspersusRows of dorsals 14 ; Oaxaca highlands and southern edge of central plateauin Michoacdn east to Veracruz imbricatus imbricatus12. Only median superciliary plate present ; preocular in contact with anteriorsupraocular ; western Chihuahua, and probably adjacent areas in Sonoraand Durango levicollis levicollisAt least an anterior as well as median, and generally a total of threesuperciliaries ; preocular rarely in contact with anterior supraocular 1313. Head scales perfectly flat (c.f. gadovii et al.) ; Oaxaca planifronsHead scales convex ; southern Coahuila to northern Guanajuato, centralDurango to San Lxiis PotosI levicollis ciliaris14. Two large loreals, one following the other (sometimes the posterior splitinto several scales) ; last two supralabials low, nearly straight-edged above(not angular) ; body and tail very elongate Uocephalus group? 20
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 369One loreal boi'doring labials, or if two one very small and split from someother scale; two posterior snpralabials nearly or quite as high as ante-penultimate labial at end of supraocular, and at least one of them (pen-ultimate) angular above antauges group? 1515. An entire postmental ; scale rows 14 16Only paired scales following mental ; scale rows 16 or more 1816. An azygous prefrontal ; range unknown ("Mexico") obscurusNo azygous prefrontal 1717. Supranasals small ; three pairs of scales between frontal and rostral ; Oaxacahighlands bocourtiSupranasals enlarged, sinjulating internasals, and forming a total of fourpairs of scales between rostral and frontal ; mountains north of Oaxacacity viridiflavus18. No azygous prefrontal ; Mount Orizaba, Veracruz antaugesAn azygous prefrontal 1919. Supranasals narrow, not at all enlarged; no ijostrostral ; Oaxaca highlandsand central Guerrero gadoviiSupranasals enlarged, simulating internasals; a postrostral (regular?);Mount Orizaba, Veracruz modestus20. A loreal segmented from lower portion of "posterior canthal," which usuallyis separated from labials ; generally two superimposed preoculars ; distinctdorsal bands; belly and tail mottled with gray and with Mack flecks;central Veracruz, foothills (not at high elevations) liocephalus ophiurusSeldom a loreal segmented from lower portion of "posterior canthal," whichrarely is not in contact with labials ; generally one preocular ; bands presentor not ; belly mottled or with black flecks but not both 2121. Dorsals 46 to 52, interparietal to posterior margin of thighs ; dorsal bandsdistinct ; no black marks whatever or ventral surfaces ; belly mottled (gray)in adults; northern San Luis Totosi through most of Coahuila to south-western Texas liocephalus infernalisDorsals 54 to 59 ; dorsal bands very indistinct in adults ; belly not mottled,generally uniform, sometimes with black flecks which may form brokenlongitudinal lines; central plateau, southern Puebla to Guanajuatoliocephalus liocephalus4.?AN UNNAMED CELESTUS FROM MEXICO, WITH A KEY TOMAINLAND SPECIES OF THE GENUSMaterial recently acquired during my tenure of the Walter Rath-bone Bacon Traveling Scholarship of the Smithsonian Institutionhas, in conjunction with other specimens in the United States Na-tional Museum, thrown considerable light upon the status of theMexican species of Celestv^. In the material available and alsothat reported in the literature, only two Mexican forms can be dis-tinguished. One of these has not been named, while the other hasreceived three names.There is another name in the literature, Euprepis microceplialusHallowell,^* which was placed in Diploglossiis (includes Celestus) by
" Proc. Acad. Nat. Sci. Phlladelphiaa, vol. 8, p. 155, 1856 (perhaps piiblisherl 1857) ; alsoTrana. Amer. Philos. Soc, ser. 2, vol. 11, pp. 79-80, 1860 (reprints perhaps appeared 1857).
370 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92Boulenger.^^ Hallowell's species was based on a single specimen saidto be from "Mexico," collected by Keating. It is obviously not thesame (by description) as the two species of Celestus now known fromMexico, however, and moreover Hallowell cites Scincus ventralis[= Oerrhonofus Uocephalus] as a synonym. The question of itsidentity was submitted to Dr. E. R. Dunn, who very kindly suppliedconsiderable additional data on the type. He states that it is not anAnguid, as Diploglossus^ but belongs to the Scincidae. The type isnot in good condition (portions of the head mutilated), but appearsto belong to a non-American genus, probably Dasia. It may possiblybe one of the original series of three specimens of Peale and Green'sScincus ventralis^ since only two of them are now present in the serieslabeled as the cotypes, and thus would be explained Hallowell's cita-tion of /Scincus ventralis as a synonym of his species. It is alsopossible that a confusion of specimens occurred, the originalGerrhonotus being exchanged for the present type of microcephalus,which Hallowell erroneously thought was one of Peale and Green'scotypes. One of these alternatives must be true: Either a peculiarskink, unknown except by the type of microcephalus, occurs in Mexico,or else some shift of specimens occurred in the Philadelphia Academycollections between 1830 and 1856. Dr. Dunn, as well as I, favorsthe latter alternative. Regardless of the provenance of the type,however, it is apparent that it is not a Biploglossus or a Celestus.CELESTUS ENNEAGRAMMUS (Cope)Siderolamprus enneagrammus Cope, Proc. Acad. Nat. Sci. Philadelphia, 1860,p. 368 (Jalapa, Veracruz).Diploglossus steindachneri Cope, Proc. Acad. Nat. Scl. Philadelphia, 1864, p.179 (Orizaba, Veracruz).Diploglossus chaljihacus Cope, Proc. Acad. Nat. Sci. Philadelphia, 1866, p. 321(Mount Orizaba, 4,000-6,000 feet, Veracruz).The 6 specimens available include the type of steindachneH(U.S.N.M. No. 6342) ; the type of chalyhaeus "(U.S.N.M. No. 6603)
;
two small specimens from Tequeyutepec, 7 miles above Jalapa, Vera-cruz (U.S.N.M. Nos. 113524-113525), which duplicate in detail thecolor description of enneagrammus^ and which are practically topo-types; a juvenile from Totontepec, Oaxaca (U.S.N.M. No. 46651);and an adult from "Tehuantepec," Oaxaca (U.S.N.M. No. 30189).Diagnosis.?The young specimens have a broad, dark brown lateralstripe on each side, extending from the snout to the sides of the baseof the tail. Bordering these stripes medially is a fine light line,continuous around snout and passing through the outer supraoculars.
? Cat. Liz. British Mus., vol. 3, p. 504, 1887.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 371A similar, fine light line extends along the middle of each of theseven median dorsal scale rows, but does not extend onto head; theremainder of the back, between the light lines, is dark brown. Theventral surfaces are bluish, and the tail is bright blue (in life). Thetwo specimens from near Jalapa, and the one from Totontepec,Oaxaca, form the basis of this description; they do not differ inpattern in any respect. The largest measures 45.2 mm. snout to vent.The three adults or subadults (smallest 65 mm. snout to vent) retainthe broad, dark lateral stripes as in the young, but the light lineson the dorsal surface increase in width at the expense of the interven-ing black, until the whole dorsal surface is light, with longitudinalstreaks of black on the edges of the scale rows. In the smallest adultthe black lines are continuous, but very narrow; in a larger adult(87.5 mm.) the black lines are broken and appear as rows of elongatespots on the edges of the scale rows; the other adult (89 mm.) is de-scribed by Cope: "Sides of head and body with limbs, black;* * * Dorsal region for a width of seven and two half rows ofscales olive brown, the edges of each row blackish and forming nar-row, imperfect lines * * * (this specimen, the type of chalyhaeus^has now lost all markings)."All specimens have rather short snouts ; posterior edge of first labialeven with middle of naris; supralabials to posterior edge of subocular7-7 in 4, 7-8 in 1 (U.S.N.M. No. 113525), 8-8 in 1 (U.S.N.M. No.113524) ; supralabials to below middle of eye 1 less than to posterioredge of subocular; 2 superimposed postnasals, except in 2 (U.S.N.M.Nos. 6603, 113525), in which the lower is fused with the first loreal;an anterior loreal following postnasals in all ; median loreal very vari-able, small or fused with adjacent scales (distinct in 1; fused wnth lat-eral prefrontals in 2 ; fused with anterior loreal in 1 ; and in 2, fusedwith lateral prefrontal on one side, with anterior loreal on other) ;distance between anterior and posterior loreals considerably less thanlength of either ; 2 superimposed posterior loreals on one side in 1, onboth sides in another, 1 posterior loreal in the remainder; 1 preocular^separated from labials on one side in 1, on both sides in another ; sub-oculars 1-1 in 1, 2-2 in 3, 2-3 in 1, indeterminate in 1 ; postoculars 3-4in 1, 4-4 in 2, 4-5 in 1, 5-5 in 1 ; supraoculars 5-5 in all ; outer supra-oculars 3-3 in all ; superciliaries 6-6 in 1, 7-7 in 3, 8-8 in 1, indeterminatein another; frontal anteriorly in contact with a single, azygous pre-frontal, a little less than twice as long as broad ; prefrontal in contactonly with anterior supraocular; 3 frontoparietals on each side (anextra 1 split off frontal on one side in 1) ; interparietal a little smallerthan either parietal, subequal to or a little larger than occipital; asmall scale split from interparietal in 1 separates it from frontal;
372 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92median prefrontal in contact with the anterior internasals in 1, sep-arating medially the scales of the posterior pair.Scale rows 33 in 4, 35 in 2 (Nos. 46651, 113525) ; lamellae under ter-minal digits flatter than others; lamellae on fourth toe 15-16 in 2,16-16 in 1, 17-18 in 1, 18-? in 1, indeterminate in 1.CELESTUS ROZELLAE, new speciesHolot7/pe?V.S.l!iM. No. 113526, an adult from the vicinity ofPalenque, Chiapas.Paratypes.?Two, one from Piedras Negras, Peten, Guatemala(U.S.N.M. No. 113527) ; the other (U.S.N.M. No. 62992) is from anunknown locality (from fruit ship at New Orleans, La.).Diagnosis.?Scale rows 31 to 33; frontal in contact with a singleprefrontal; latter in contact with only anterior supraocular; 8 to 10labials to posterior edge of supraocular ; first labial not extending be-yond anterior margin of naris ; median loreal as long as either of theothers, or longer; portion of rostral visible from above greater than itsdistance from second pair of internasals ; sides dark with vertical lightbars, in young and adults ; young with a broad, dorsolateral light stripeon each side, fading and indistinguishable in adults.Description of holotype.?Portion of rostral visible from above two-thirds as long (1.1 mm.) as its distance from prefrontal (1.5 mm.) ;two pairs of internasals; a large median prefrontal in contact withanterior supraocular and narrowly separated from anterior supercili-ary; width of frontal (3 mm.) two-thirds its length (4.7 mm.) ; inter-parietal subtriangular, its suture with frontal a little smaller thansuture with either anterior frontoparietal ; interparietal and occipitalsubequal in size, two-thirds size of either parietal ; 3 frontoparietals,anterior largest, median smallest ; nasal very elongate, nostril piercedposteriorly ; 2 superimposed postnasals ; 3 loreals, 2 anterior subequalin size, smaller than posterior; lateral prefrontal a little larger thanmedian loreal, separating latter from median prefrontal; a single pre-ocular, equal in size to anterior loreal ; a short anterior and long poste-rior subocular; 4^5 postoculars; 8 superciliaries; 5 primary (inner)supraoculars, 3 secondary outer supraoculars; posterior edge of firstlabial about even with anterior border of naris ; 7-8 labials to belowmiddle of eye, 8-9 to posterior end of subocular ; 1 postmental and 3large pairs of chinshields, the scales of the anterior pair in contactmedially.Ear opening small, much smaller than eye opening; 31 scale rowsaround middle of body; 21-22 lamellae under fourth toe; lamellaeunder distal phalanx larger and flatter than others.Snout to vent 84 mm. ; snout to posterior border of ear 14.9 mm.
;
foreleg 19.2 mm. ; hind leg 26 mm.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 373Color.?Head gray-brown, with a few darker brown flecks, andsome of the scales edged with dark brown; sides of body and neckdarker brown, with numerous, somewhat irregular, vertical light barsa little over a scale wide, and separated from each other by the widthof from two to three scales ; dorsal surface gray-brown as head, with
.small dark flecks scattered on many of the median scales; scales justabove the dark sides with very few flecks; tail a little lighter thanbody, with longitudinal flecks of brown on the centers of the scales,particularly prominent on the two median scale rows. Ventral sur-faces of body and tail bluish, except for the preanal region, a narrowarea across chest, posteroventral surfaces of limbs, and midventralsurface of tail, which are white.YaTiation.?ViXQ adult paratype (U.S.N.M. No. 62.992) is markedlike the holotype. The juvenile from Piedras Negras, however, whichmeasures 44 mm. snout to vent, is somewhat different. The sides ofthe body are dark and with vertical light bars as in the adults; thelight bars are not quite so broad and a little more broken into spots
;
they extend somewhat into the dorsal region. A very broad, blackband extends from the snout along the middle of the back onto thebase of the tail ; it covers three and two half-scale rows on the middleof the back, and on the head involves nearly all of the prefrontaland parietals. This band is separated from the lateral band by abroad light line extending from snout through the lateral supraocularregion to the tail; these cover two and two half-scale rows at themiddle of the body, and are greenish in the temporal region, creamon snout, and of a golden tint over most of the body. The laterallight streaks are very pale blue. Ventral surface of body rather brightblue, lighter on chin and on limbs.There is but little variation in scutellation. In 1 the anterior super-ciliary is narrowly in contact with prefrontal on one side ; the occipitaland interparietal are half the size of the parietals in 1 ; median lorealfused with scale above it on one side in 1, but the distance between theanterior and posterior loreals still greater than the length of either;7-8 superciliaries in 1; 7-8 supralabials to below middle of eye in 1,8-9 in other (8-9, 9-10, respectively, to posterior border of subocular)scale rows 31 in 1 (No. 113527), 33 in other; lamellae under fourthtoe 23-23, 26-?Remarks.?The holotype was discovered during the day by mywife, for whom it is named. It was rapidly running up the trunk ofa small tree near an open spot in a wooded area.The species differs from enneagrainiiius chiefly in the elongate,flattened snout and different pattern. The lengthening of the snoutis expressed in the larger size of the medial loreal, the elongation ofthe nasal, and the enlargement of the rostral.
374 PROCEEDINGS OF THE NATIONAL MUSEUM vol.92KEY TO MAINLAND CELESTUS
1. Three prefrontals in contact with frontal " ; green above, lighter on sides(adult) montanusOnly one prefrontal in contact with frontal 22. Two loreals ; prefrontal in contact with two supraoculars ; 6 labials to belowmiddle of eye; a broad dorsolateral light stripe; sides uniform dark, inyoung bivittatusThree or more loreals (the median may be fused with another scale, the lateralprefrontal) , or, if only two, prefrontal separated from second supraocular. 33. Postnasals followed by two pairs of superimposed loreals ; "frontal twice aswide as long; four external and four internal supi'aorbitals" cyanochlorisPostnasals followed by two or three loreals, not paired ; frontal much longerthan broad ; five inner supraoculars, three outer 44. First labial reaching to middle of naris ; distance between anterior and pos-terior loreals less than the length of either ; no vertical light bars on sidesin young or adults enneag-rammusFirst labial reaching to anterior border of naris ; distance between anteriorand posterior loreals practically as great as, or greater than, the length ofeither; vertical light bars present on sides in young and adults rozellae5.?NEW XANTUSIID LIZARDSThe collections obtained in Mexico through tlie aid of the WalterRathbone Bacon Traveling Scholarship in 1938 to 1940 have revealedthe existence of two previously unknown species of Gaigeia^ and anundescribed race of Lepidophy7na. Specimens previously collectedfor the United States National Museum by E. W. Nelson and E. A.Goldman include still another unnamed race of Lepklophyma^ mak-ing a total of four species and subspecies of each genus in Mexico.GAIGEIA DONTOMASI, new speciesHolotype.?U.S.N.M. No. 111473, an adult female from Lachiguiri,Oaxaca, at 7,100 feet, collected by Thomas MacDougall, January20, 1940.Paratype.?No. 111474, a topotype, same collection data as theholotype.Diagnosis.?Dorsal whorls three to a caudal segment, ventralwhorls two; scales on body of nearly equal size; no distinctly en-larged, keeled scales on thigh ; four or five rows of granules middor-sally; three temporals, the anterior half as large to as large as theposterior; enlarged scales in paravertebral rows separated from eachother by an average of two scales.Description of holotype.?Head and body somewhat flattened;length of portion of rostral visible from above considerably greaterthan its distance from frontonasal ; nasals in contact medially behind
^^This is also true of nuchalis Boulenger (Proc. Zool. Soc. London. 1898, p. 920, pi. 56,fig. 1 ) from an unknown locality.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 375
rostral ; frontonasal in contact with anterior loreal ; three prefrontals,the median slightly smaller than others and narrowly separated fromfrontonasal; two frontals, their median suture less than half theirmaximum length; interparietal hexagonal, longer than and as largeas either parietal; temporals three, the anterior and posterior sub-equal in size and less than a fourth size of parietals; median tem-poral a little larger than or subequal to a parietal, very narrowlyseparated from frontal; two nasals, the nostril pierced in anterior;two loreals, the anterior not quite two-thirds size of posterior; twopreoculars, lower larger, triangular; three suboculars, three postocu-lars and three superciliaries, all small; eight supralabials, the fifthbelow middle of eye ; granular temporals a little larger than nuchals,somewhat irregular in size, becoming larger anteriorly; a row ofsmall auricular lobules bordering ear.A large mental, with a considerably larger labial border thanrostral; three large pairs of scales (chinshields) following mental,the first two in contact, the posterior separated by about four scales(three to five) ; gular scales small, rounded, not imbricating, becom-ing flat and a little larger anteriorly ; a rather well defined gular foldpreceding arms, terminating on sides of neck, ventral ly involvingsmaller scales than occur anteriorly.Sides and back with small, nonimbricating scales of nearly equalsize, most conical or bluntly keeled; four or five middorsal rows ofsmall, uniform granules, bordered on either side by a series of ratherwidely spaced, somewhat enlarged scales separated from each othergenerally by two scales; on dorsolateral surface are similar, scat-tered, slightly enlarged scales; sides with granules of uniform size.Thirty-two transverse rows of enlarged, flat, juxtaposed, ventralscales from gular fold to anus; 10 longitudinal rows of ventrals atmiddle of belly; 4 large, subequal preanals, and in addition theposterior pair bordered on either side by a scale half as large.Dorsal surface of foreleg with small, conical tubercles, smaller onlower foreleg; ventral surface similarly protected, except that thesmaller scales are on the upper foreleg, the large on the lower ; lamel-lar formula for fingers 10-15-15-10-7.Dorsal surface of thigh with nearly uniform, conical tubercles,shank with scales of irregular size, the larger ones bluntly keeled;ventral surface of hind leg with flat scales, largest anteriorly onshank; femoral pores 10-10; the pore series terminating medially ata large, flat scale; lamellar formula for toes 16-23-18-12-8.Dorsal caudals feebly keeled, subcaudals smooth ; on dorsal surface,every third whorl of scales slightly enlarged; first whorl of eachcaudal segment (i. e., that which follows the whorl of enlargedscales) incomplete ventrally.
376 PROCEEDINGS OF THE NATIONAL MUSEUM vol.92Snout to vent 50 mm.; tail 61 mm. (regenerate); foreleg (fromaxilla) 15 mm. ; hind leg (from base of tail) 21 mm.Color.?Dorsal surface light gray, lighter on head; no marks ondorsal surface of latter; sides of head dark brown (black), withnumerous, large, light areas in labial and lower temporal region;centers of supralabials dark; chinshields and mental with large darkareas toward border, and a smaller dark area medially ; throat with afew small flecks of black ; posterior temporal region and sides of bodywith small, scattered, round, light spots bordered by black; these spotsform a more or less definite paravertebral and dorsolateral row on eachside, and in these rows their dark borders expand anteriorly andposteriorly and become continuous; middorsal area unmarked; tailwith similar, scattered light spots (which occupy more than half thetotal caudal surface), but here they are rectangular; large, light,dark-bordered ocelli occur on the limbs as well as on tlie body; ventralsurface of body light, with some dark suffusion on the anterior edgesof the belly scales, especially prominent laterally; subcaudal surfacespotted (except midventrally) as dorsal surface.Variation.?The paratype (male) is essentially similar to the holo-type. The anterior temporal is half the size of the posterior; themedian prefrontal is in contact Avith the frontonasal. There are 10-10femoral pores.Remarks.?The specimens are named for tlioir collector, ThomasMacDougall, in whose company we spent several very pleasant weekson the Isthmus of Tehuantepec. They were found under fallen logsin a pine forest. GAIGEIA RADULA, new speciesnolofypc.?U.^.Yl.M. No. 111472, an adult female from San JoseManteca, 5 kilometers from San Carlos Yautepec, Oaxaca.Diagnobis.?Dorsal whorls three to a caudal segment, ventral whorlstwo; scales on body irregular in size, with many, closely placed, en-larged, keeled scales ; thigh with enlarged, keeled scales ; two or threerows of granules middorsally; three temporals, the anterior a thirdsize of posterior; enlarged scales in paravertebral rows separated fromeach other by an average of one scale.Desanftion of holotype.?Similar to dontomcsi in general featuresof cephalic scutellation ; frontals rather narrowly in contact ; a short,vertical series of somewhat enlarged scales preceding ear.Dorsal scales mostly keeled and conical ; 2 or 3 rows of granules mid-dorsally, bordered on either side by a series of enlarged, keeled scalesgenerally separated from each other by 1 small scale; scales on sides ofbody irregular; scattered, enlarged, keeled, closely placed tuberclespresent; many of dorsal scales of thigh enlarged, keeled; femoralpores 10-10; ventral scales 32, in 10 longitudinal series at middle ofbelly; lamellar formula for toes 14-20-16-10-6.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 377Color.?Dorsal color dark gray; sides of head darker, with vaguelight areas ; a dark stripe through upper temporal region ; dark markson labials, mental and chinshields as in dontomasi; a body patternfaintly discernible, similar to that of dontomasi.ReTnarks.?This species is obviously a close relative of dontomasi;it is with some hesitation that I have held the present specimen as dif-ferent from the other two. The differences between radula and donto-masi are rather striking, however, in scutellation of the body, as canbe seen by a comparison of the diagnoses of the two.LEPIDOPHYMA SMITHII TEHUANAE, new subspeciesHolotype.?V.^.'^.'M. No. 111488, from Cerro Arenal, 30 kilometerswest of Tehuantepec, Oaxaca.Paratypes.?E\%\ej\. U.S.N.M. No. 111489, EHT-HMS 28136,Tres, Cruces, Oaxaca; U.S.N.M. No. 111490, El Limon, Oaxaca;U.S.N.M. Nos. 111491, 111492, La Concepcion, Oaxaca (all between30 and 50 kilometers west of Tehuantepec) ; U.S.N.M. No. 46687,Santa Efigenia, Oaxaca ; U.S.N.M. Nos. 46997 to 47000, near Tehuan-tepec, Oaxaca; and U.S.N.M. No. 48105, mountains near SantoDomingo, Oaxaca.Diagnosis.?Three dorsal and two ventral whorls of small scalesseparating the whorls of enlarged scales on tail ; femoral pores 13 orless; no median frontonasal; very young with pink tails.Description of holotype.?Head scutellation typical, except: Nomedian frontonasal; anterior of the three enlarged temporals small,separated from labials by a narrow row of granules and a row ofsomewhat enlarged scales; scales in posterior temporal region ir-regularly enlarged, some of the larger ones arranged in a diagonalrow a little in front of ear; no enlarged scales adjoining enlargedtemporals; auricular lobules prominent; granules in gular regionrelatively large, about nine in the vertical diameter of ear opening(at about middle of throat).Enlarged tubercles on sides of body flattened on their posteriorfaces, arranged in vertical rows in which seldom more than two ofthe tubercles are separated on both sides from the adjacent enlargedtubercules ; generally four rows of granules middorsally between para-vertebral rows of enlarged tubercles.Ten longitudinal rows of ventrals, 37 rows from gular fold toanus; 9-10 femoral pores; 75 mm. snout to vent. Three rows ofsmall, keeled scales dorsally on tail, separating the whorls of en-larged scales ; scales of whorl following the whorl of enlarged scalesmuch reduced in size toward base of tail, proportionately equal toothers only distal to about middle of tail ; the latter whorl is droppedon the ventral surface, leaving 3 whorls to a segment.
378 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92Color.?Dark brown above, with small, light ocelli on body arrangedin paravertebral, dorsolateral, and lateral rows on each side; bellyunmarked; gular region feebly mottled; dark marks present in thecenter of each chinshield (infralabial) and of mental.Variation.?The femoral pores in the 8 adult paratypes vary from10 to 13 (10, 6; 11, 2; 12, 5; 13, 2). The 3 juveniles (largest 29 mm.snout to vent) had bright pink tails in life; after a year of preserva-tion they appear of very light flesh color, with vague reticulations ofdark pigment.All specimens lack a median frontonasal, and largely on the basisof this character the specimens from the Tehuantepec area are dis-tinguished from typical smithii. This scale is regularly present insmithii., as shown by a large series of 127 specimens from variouslocalities in southern Chiapas near Escuintla. The young smithii^moreover, many of which were seen in life, do not have pink tails, butrather the tails are dark, as are the bodies.LEPIDOPHYMA SMITHII OCCULOR, new subspeciesHolotype.?U.S.N.M. No. 47133, from Jalpan, Queretaro, collectedby Nelson and Goldman.Paratypes.?Tvco topotypes, U.S.N.M. Nos. 47134-47135.Diagnosis.?Three dorsal and 2 ventral whorls of small scales sepa-rating the whorls of enlarged scales on tail ; femoral pores less than14 (in known specimens, 10-11) ; no median frontonasal; young un-known; scales anterior to ear very small and uniform, except for arow of enlarged tubercles adjacent to enlarged 'temporals; whorlson tail relatively poorly differentiated; most of tubercles on sidesof body separated from each other; about 12 granules near middleof throat contained in vertical diameter of ear opening.Description of holotype.?Head scutellation typical, except: Nomedian frontonasal; anterior of the 3 enlarged temporals small, sep-arated from labial by a single large scale nearly as large as labial;scales in front of ear (anterior to prominent auricular lobules) verysmall and even in size, except for a row of enlarged ones adjoiningthe enlarged temporals; granules in gular region very small, about12 (at middle of throat) in the vertical diameter of ear opening.Enlarged tubercles on sides of body flattened on their posteriorfaces, arranged in vertical rows in which practically all the enlargedtubercles are separated from each other ; generally five rows of gran-ular scales midclorsally between paravertebral rows of enlarged scales.Ten longitudinal rows of ventrals; 37 rows from gular fold to anus;10-11 femoral pores ; 93 mm. snout to vent.Three whorls of small, keeled scales dorsally on tail, separating thewhorls of enlarged scales; scales of whorl following the whorl of
MEXICAN HERPETCLOGICAL MISCELLANY?SMITH 379
enlarged scales much reduced in size toward base of tail; this whorldropped on ventral surface, leaving three whorls to a segment.Color.?Dark brown above, with large, anastomosing light spots;head light; belly, throat, and tail unmarked; chinshields with feebledark marks.Vanation.?The two paratypes have 10 femoral pores on each side(uncertain on one side of one). Both lack a median frontonasal, andthe temporal scales anterior to ear and gular scales are small, as inthe type.Comparisons.?This subspecies is similar to tehuanae in the absenceof the median frontonasal, but it differs from that subspecies as wellas from smUhii in the small size of the temporal and gular scales,presence of a row of enlarged tubercles next to the enlarged tem-porals, and in the reduction in size of the scales in the enlarged tailwhorls.The 3 subspecies of smithii hold in common a low femoral porecount, a maximum of 4 (1 incomplete) dorsal and 3 ventral whorlsto a caudal segment, and in these characters differ rather widelyfrom -flavomaculatum., which usually has 14 or more femoral poresand a maximum of 5 whorls (2 incomplete) to a caudal segment.In 17 typical specimens of favomaculatum (Guatemala and Mexico),the median frontonasal is absent in 1, the femoral pores are 15 to 22(15, 1; 16, 4; 17, 4; 18, 7; 19, 1; 20, 3; 22, 1), and in all the additionaldorsal whorls of caudal scales are present even near the base of thetail. In 12 /. ohscurum from Tela, Honduras, to Panama, the medianfrontonasal is absent in 3, and the femoral pores vary from 13 to 19(13, 1; 14, 5; 15, 10; 16, 2; 17, 4; 18, 2; 19, 1) ; in the 2 Panama speci-mens the tail is as in typical favoinaculatum, while in the others theadditional whorls become evident only distal to the middle of the tail.These data seem to indicate that there is a closer relationship be-tween smithii^ tehuanae., and occulor than between any of them andflavomaculatum or ohscurum: For this reason the former three areassumed to be races of one species, while the latter two belong toanother. KEY TO MEXICAN LEPmOPHYMA AND GAIGEIA
1. Distinct, vertical rows, separated from each other by granular areas, of well-differentiated, enlarged, keeled scales on sides of body Lepidophyma 2No distinct, vertical rows of enlarged scales on sides of body Gaigeia 52, All (except basal) whorls of enlarged scales on tail separated from eachother dorsally by 4 rows of scales ; femoral pores 15 to 22 ; median prefrontalnormally present, sometimes absent; Atlantic slopes. Tabasco into Guate-mala flavomaculatum flavomaculatumAll whorls of enlarged scales on tail separated from each other dorsally by3 rows of scales (rarely feeble evidence of a fourth row) ; femoral poresless than 14; median prefrontal present or absent 3
380 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 923. A median prefrontal ; Pacific slopes, Chiapas into Guatemalasm.ithii smithii "No median prefrontal 44. Scales in posterior temporal region (anterior to ear) minute and very uniformin size, except for a series of relatively large, projecting auricular lobulesand a row of larger scales beside the upper temporals ; whorls on tail rela-tively little differentiated ; Atlantic slopes, probably from southern San LuisPotosi to northern Veracruz ; type locality Jalpan, Quer^taro.smithii occulorScales in posterior temporal region (anterior to ear) larger, irregular in size;auricular lobules poorly defined or absent; no scales bordering uppertemporals; tail whorls strongly differentiated; Pacific slopes, Isthmus ofTehuantepee smithii tehuanae5. All whorls on tail complete, none restricted to dorsal surface gaigeaeSome of scale whorls on tail restricted to dorsal surface 66. Only one row of scales on ventral surface between the whorls of en-larged scales (i. e., every third whorl restricted to dorsal surface) 7Two proximaliy, three distally, rows of scales on ventral surface between thewhorls of enlarged scales sylvatica7. Numerous, very closely approximated, enlarged, keeled scales, separated bysmall granules, present on sides of body ; two or three rows of granules invertebral region radulaDorsal scales practically uniform in size ; four rows of granules in vertebralregion dontomasi6.?THE MEXICAN SUBSPECIES OF DRYMOBIUS MARGARITIFERUSI first observed that very easily recognizable differences occur inMexican Dnjmohms margaritiferus when my wife and I were collect-ing in the vicinity of La Esperanza, Chiapas, in April and May of1940, as guests of Mr. and Mrs. Eizi Matuda. Although we had col-lected Drymobius in most of the provinces in Mexico where it occurs,and had never before had any hesitancy whatever in identifyingmargaritifer-us as such at first sight, the specimens we collected insouthern Chiapas were so different that at first we did not recognizethem at all, and only after capturing several sepcimens did we realizethey might be close to the familiar Drymobius margaritife')ms ofareas to the north.Subsequent study of these and other Mexican specimens revealed nosignificant differences in scutellation, but did verify the remarkabledifferences in color and pattern that we had observed in the field.
"An overlooked synonym of smithii is tbe monotypic AMeistops guatemalensis Miiller,Verh. Naturf. Ges. Basel, vol. 6, pp. 390-398, pis. 1-2, 1878. It is also to be noted thatthe "Tehuantepee" types may or may not be typical smithii as described and figured byBocourt. This author mentions six specimens ot his new form, four from "Tehuantepee"(Sumichrast) and two from the west coast of Guatemala. However, the species is bothdescribed and figured with a median frontonasal, and no specimens are mentioned withoutone. It must be concluded that the "Tehuantepee' specimens either are not from thevicinity of Tehuantepee city but from the extreme eastern part of the Isthmus (near or inChiapas), or that the scale is actually absent, since Bocourt, paying no great attention tothe presence or absence of it, neglected to specifically describe its nature. In view oftho doubt concerning the nature of the "Tehuf.ntepec" types, I believe it best to restrictthe name to the form best characterized by his description and figure, and to the eotypefrom Guatemala, which beyond (|uestion belongs to the same race as the large seriesavailable from the vicinity of La Esperanza, Chiapas.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 381The southern Chiapas specimens are characterized chiefly by theirdull color, which differentiates that population from all other rnarga-rltifer-us of Central America, Mexico, and the United States. To thisthe name occidentalis Bocourt is applicable. In mature specimens ofthis race the light central spot of each dorsal scale (characteristic ofthe whole species) is gray-brown, diffuse, poorly defined, and stippledwith darker. Young specimens are the same, except that the lightareas are better defined. In all other mature specimens of the species,from Costa Rica to Texas, the light central spot is blue, or yellow toorange, or is partly blue and partly yellow or orange. The differenceis striking, even in many long-preserved specimens (unless badlydiscolored by formalin).Correlated with this color difference in occidentalis are two readilydiscernible differences in pattern. The most uniform difference, butperhaps less readily defined in words and also perhaps varying withage, is the complete absence (in mature specimens) or poor definition(in young specimens) of the dark area on the side of the head.Of all other specimens of the species this dark area is very charac-teristic, is well defined, and (as a key basis of comparison) is muchdarker than the general tone of the median dorsal nape area (equalin occidentalis). A more easily definable pattern difference of occi-dentalis is the complete absence of black edges on the subcaudals;the ventral surface of the tail is white, totally immaculate. Manyspecimens from the Atlantic slopes of Mexico and Texas have beenchecked for this character, and without a single exception all havethe posterior edges of the subcaudals black. Frequently the bellyscales are black-edged also (not or only laterally in occidentalis).This character, therefore, completely separates Atlantic coast Mexi-can and United States margaritiferus from occidentalis, but the char-acter fails when applied to Pacific coast specimens from Tehuantepccnorth, in which is found the proportion of one white-tailed specimento two banded-tailed specimens, with some specimens arbitrarily allo-cated to one category or the other.Accordingly, other differences were sought, and it was then dis-covered that Pacific coast specimens uniformly differ from Atlanticcoast specimens in having black the entire border, anterior as well asposterior, of the middorsal scales (not the extreme lateral scales).In Atlantic coast specimens the posterior border of the middorsalscales is black, but the anterior borders (from center) are blue; indiscolored and young specimens, of course, the anterior borders donot appear blue, but are readily discernible as lighter and well dif-ferentiated from the jet black border of the posterior edge. Thedifference is not uniformly discernible in the pattern of the lateralscales, but is well defined on the middorsal scales.469013?42 3
382 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92Application of tliis character to occklentaJ'is does not yield com-pletely satisfactory results, however, for the black tips of the mid-dorsal scales shade more or less gradually into a gray, stippled posteriorportion. Nevertheless, the four characters above described (1, dorsalcolor, light spots; 2, black areas on sides of head; 3, bands under tail;4, individual pattern of middorsal scales) do serve to differentiatethree races in Mexico : occidentalism in southern Chiapas and southernGuatemala; viavgaritijerus^ on Atlantic slopes from southern Texasinto Central America ; and an unnamed form described below as jistu-losus, occurring from the Tehuantepec area nortliward on Pacific slopesto southern Sinaloa. It is of great interest that specimens fromTonala, Chiapas, are neither fsfnios-us nor oceidentalis, but typical
'iiiargaritiferus. This distribution corresponds more or less with thatof Scelopoims v. variabilis. It is the only area in Mexico where typicalTJiargaritiferus occurs on Pacific slopes.In studying differences in Mexican mdrgaritiferus I have largelyneglected Central American specimens, some of which I realize donot conform to the present definition of typical margaritiferus (sub-caudals not uniformly black-edged) . Their dorsal color prevents anyexcept those from southern Guatemala from inclusion with occidentalismand their separation geographically from fistulostis, as well as theindividual scale pattern (like inargarit^feims) , makes unwise associa-tion of any specimens with that race. Pending further studies, prob-ably all Central American specimens (except oceidentalis) should bereferred to typical margaritiferus, with which i\\By agree in mostrespects.The Mexican specimens I have examined (about 100) can be identi-fied easily by the accompanying key, with the exception of somejuveniles (discolored) and of certain adult specimens from theTehuantepec area, which also unfortunately are discolored by longpreservation in formalin. Specimens thus discolored cannot always bedefinitely identified, and are best named by geographic probability.KEY TO MEXICAN SUBSPECIES OF DRYMOBIUS MARGARITIFERUS
1. All or most of posterior edge of siibcaudals dark or black 2All or most of posterior edge of subcaudals white (the subcaudal surfaceunmarked) 82. Anterior edges of median dorsal scales blue (in adult specimens), white (fadedspecimens) or gray (young specimens), the color well differentiated froma black tip of scale margaritiferusAnterior edges of median dorsal scales black, the color completely surroundinga light central spot fistulosus3. Sides of head in temporal region no darker than general tone of dorsal surfaceof nape; light spots in centers of scales diffuse, gray-brown, stippledoceidentalis
MEXICAX HERPETOLOGICAL MISCELLANY?SMITH 383An elongated dark (black) area on either side of head behind eye, much darkerthan general tone of dorsal surface of nape ; light spots in centers of scaleswell-defined, blue or tinged with orange, not or very slightly stippledfistulosusDRYMOBIUS MARGARITIFERUS MARGARITIFERUS (Schlegel)Herpctodrijas margaritiferus Schlegel, Essai Phys. Serp., vol. 2, p. 184, 1837(New Orleans, Louisiana, by error; here restricted to Veracruz, Veracruz).Diagnosis.?Subcaudals black-edged posteriorly; a distinct dark{black) area in temporal region, much darker than any other part ofhead, darker than general tone of body anteriorly; posterior edge ofeach middorsal scale black, the anterior (concealed) edges blue (inadults; white in faded specimens, gray in young), middle yellow orlight orange (white in faded specimens, light in young)./Specimens examined.?Forty-six from Mexico, 6 from Texas, vari-ous from Central America. Mexican localities represented by speci-mens examined are in the states of Tammdipas (Alta Mira, AntiguoMorelos, 7 miles west of Victoria, Hda. La Clementina nr. Forlon),San Luis Potosi (Jilitla), Veracruz (Catemaco, Cuautlapan, Mina-titlan, Mirador, Orizaba, San Rafael, Potrero Viejo, Tuxpan) , Tabasco(Tenosique), Chiapas (El Salto and San Juanito, nr. Palenque;Tonala), Campeche (Campeche), and Yucatan (Chichen Itza, Puerto
\ Morelos).
1 Range.?Atlantic slopes below about 4,500 feet, from extreme
I fsouthern Texas into Central America ; also western Chiapas, Pacific
j slopes.
j DRYMOBIUS MARGARITIFERUS OCCIDENTALIS Boconrt
I
DrymoHus margaritiferus occndentalis Bocotjrt, Miss. Sci. Mex., Rept., p. 718,
I 1890 (Volcan Atitlan, Guatemala).
i Diagnosis.?Subcaudal surface white, completely unmarked ; sides ofhead in temporal region little if any darker than median parietalregion, about same shade as general tone of dorsal surface on nape;central spots of dorsal scales gray-brown, diffuse, stippled ; black tips
I of dorsal scales shading into gray posterior color.JSpecimens examined.?Nine, all from the vicinity of La Esperanza(near Escuintla), Chiapas.Range.?Pacific slopes of southern Chiapas and of Guatemala.DRYMOBIUS MARGARITIFERUS FISTULOSUS, new subspeciesEolotype.?U.S.N.M. No. 51480, female, from Miramar, Nayarit,collected by J. C. Thompson in 1913.Paratype.?V.^.'^M. Nos. 31480-31483, 56163, Colima; No. 30484,Tehuantepec, Oaxaca; No. 46545, Puente de Ixtla, Morelos. EHT-HMS No. 4607, Paso del Rio, Colima; Nos. 4608, 5363-5365, Hda.
384 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92El Sabino, nr. Uruapan, Michoacan; No. 4613, Cuernavaca, MorelosjNo. 5366, Ocotito, Guerrero; No. 23627. Tierra Colorada, Guerrero.Diagnosis.?Subcaudals usually black-edged, but not always (about65 percent) ; a distinct dark (black) area in temporal region, muchdarker than any other part of head, darker than general tone ofbody anteriorly ; entire border, anterior as well as posterior, of mid-dorsal scales black, a small central area blue or yellow.Description of hoJofype.?Nine supralabials, fourth and fifth en-tering orbit; 9-10 infralabials; ventrals 146; anal divided; tail in-complete.Anterior part of head (in front of frontal) light brown, posteriorpart bluish; a large dark brown spot including temporal region,upper part of posterior supralabials, sides of parietals, and most ofsupraoculars; body black, with a light spot in the center of eachscale; anterior spots blue, and with a small yellow center, posteriorspots mostly yellow; light spots not reaching anterior border ofscales except on lateral anterior scales, on all others restricted tocenters of scales by a complete black border; a dark line, dimmertoward venter, marking each ventral suture ; similar lines, but betterdefined, marking subcaudal sutures.Specimens examined.?Besides the type series (16), 17 (discoloredby formalin) from the vicinity of Tehuantepec, Oaxaca.Range.?Pacific slopes, southern Sinaloa to the region about Tehuan-tepec; does not reach Tonala, Chiapas.7.?NOTES ON MEXICAN IMANTODESSpecimens of Imantod.es in the Museum of Comparative Zoology,United States National Museum, and the E. H. Taylor-H. M. Smithcollection, examined in preparation for a checklist and key to Mex-ican snakes, do not conform with current nomenclature. A numberof changes appear necessary, and so that at least some of thosewhich involve Mexican species may be available for citation, the fol-lowing arrangement is presented.I am much indebted to Dr. E. E.. Dunn for very generous assistanceand the benefit of his observations on Central American Imantodes,to Arthur Loveridge for the loan of specimens in the Museum ofComparative Zoology, and especially to Dr. E. H. Taylor forloan of specimens and for the photographs here reproduced.IMANTODES CENCHOA LEUCOMELAS CopePlate 37, Figure 1.Himantodes leucomelas Cope, Proc. Acad. Nat. Sci. Philadelphia, 1861, p, 296(Mirador, Veracruz; U.S.N.M. Nos. 25035-25036.)Dipsas cenchoa rhombeata Mutxer, Verb. Naturf. Ges. Basel, vol. 7, p. 151, 1882(Guatemala).
J. S. NATIONAL MUSEUM PROCEEDINGS. VOL. 92 PLATE 37owcn
1, Imantodes cenchoa leucomelas Cope, EHT-HMS No. 5585, from Atoyac, V'eracruz.2, Imantodes latistratus Cope, EHT-HMS No. 5517, from near Cuernavaca, Morelos.

MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 385Diagnosis.?Vertebral scales 3 to 4 times as broad as adjacent scales;dorsal spots on body 31 to 40; no V-shaped light mark bifurcating
?n posterior part of parietals, the arms reaching toward posteriororner of eyes; bands rarely broken on sides of body posteriorly.Range.?Mexico north to central Veracruz, southern Chiapas, atleast parts of Guatemala.Specimens exam ined.?Twenty-two.Remarks.?This form is easily distinguished from Central and SouthAmerican cenchoa., and perhaps does not even intergrade with them.
\ The differences are of a type that could conceivably show intergra-
I dation, however, and furthermore all are members of a very definite,compact morphologic group characterized by having the vertebrals
I three times as wide as adjacent scales, or wider, and usually two labials
I entering orbit.^^It appears that 3 forms are distinct: leucomelas, with 30 to 40bands on body, bands not or rarely broken on sides of body pos-
! teriorly, no V-shaped light mark bifurcating on posterior part of
I frontal and extending toward posteromedial border of orbit; serni-
i fasciatus.^ with 40 or more bands on body, bands broken on posterior
! part of body and consisting of a small lateral spot widely separatedfrom a large spot restricted to dorsal area, and head markings as inleucomeJas; and typical cenchoa.^ with 39 or more bands on body, bandsriot or rarely broken on sides of body posteriorly, and a V-shaped
I light mark bifurcating on posterior part of parietal, usually reach-ing posteromedial border of orbit. The subspecies semifasciatus oc-curs from Nicaragua to Panama, and cenchoa from Panama to SouthAmerica. I have seen no specimens of this morphologic group fromHonduras, and cannot say which form occurs there. North of Hon-duras all specimens examined are typical and undoubted leucomelas,while south of Honduras all are semifasciatus or cenchoa.IMANTODES GEMMISTRATUS CopeIlimantodcH geinmistratns Cope, Proc. Acad. Nat. Sci. Philadelphia, 1860, p. 264(Izalco, Salvador).ffimarttodcs cenchoa clegann Jan and Sobdelli, Icon. Gen., livr. .38, pi. 2, fig. 1,1871 (Central America).Dlpsas cenchoa reticulata Mulleb, Verh. Naturf. Ges. Basel, vol. 7, p. 151, 1882(Guatemala).Leptognathiis stratissima Cope, Proc. Amer. Philos. Soc, vol. 23, p. 280, 1886(Panama).
" In a letter Dr E. R. Dunn tells me that "of 291 cenchoa from Nicaragua, Costa Rica,*nd Panama, only 16 have more than two labials in orbit. Of 188 elegans eight haveless than three labials in orbit." The proportion is higher in Mexican leucomelas, how-
??ver, making impossible the use of this character in keys : three labials enter orbit on one
?or both sides in 8 out of 18 specimens.
386 PEOCEEDINGS OF THE NATIONAL MUSEUM tol. 92Diagnosis.?Vertebrals enlarged, about twice width of adjacentscales ; usually three labials enter orbit ; bands on body numerous, 52to 65, broken posteriorly or not.Range.?Pacific coast of Chiapas, south to Panama.Specimens examined.?Four.Remarks.?This species is distinguished from all others by havingmost of the vertebral scales about twice as wide as the adjacent scales
;
they are larger than in any other species except cenchoa.The name gemmistraius has long been used for the Mexican Iman-todes with scarcely or not enlarged vertebrals. Even Cope eventuallyused it in this sense. Unfortunately, the type is missing, so it isimpossible to prove what form Cope may have originally described.Fortunately, however, the type locality is definite. It is fairly cer-tain that the name does not apply to cenchoa.^ because (1) Copedefinitely says the vertebrals are relatively small, (2) he gives 42bands on body, while 40 is the maximum for leucomelas (definitelyknown from southeastern Chiapas, therefore probably the speciesof Pacific coast Salvador, which is in the same faunal zone), and (3)he says the bands "are peculiar in being connected by a median dorsalvitta." The latter is a characteristic of many specimens of otherspecies of Imantodes., but not of cenchoa.I conceive that gemmistratus and splendidu^ (see below) are mem-bers of a single morphologic group (having similar ventral and caudalscale counts, similar pattern), and accordingly that one does not occurwith the other. If this is true, then the species with small dorsals{gemmistratus iiuct.= splendidus) cannot occur in the same territorywith the species with larger dorsals {eJegans anct.^ gem/misfrafus)
.
Since the only specimens now known from the faunal area that includesIzalco, type locality of gemmistratus, are two ^^ which have the verte-bral scales enlarged (as in elega-ns auct., I believe the name gemmi-stratus must be applied to that species. Unfortunately it is an oldername than elegans, which it replaces.While gemmistratus is very uniform in character of the vertebralscales, extensive variation in pattern occurs. Panama and Costa Ricaspecimens have the bands complete on the posterior part of the body,as well as anterior, while Nicaragua and Chiapas specimens have thembroken posteriorly. The ventrals and caudals are usually low (as inMexican splendidus), and vary from 220 to 237 (ventrals) and 114 to146 (caudals) . The exception is a small Nicaragua specimen with 251ventrals and 155 caudals (U.S.N.M. No. 25248).
i?I>a Esperanza, Chiapas (U.S.N.M. No. 110521) and Pacific Coast of Guatemala (Brit.Mus.).
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 387IMANTODES GRACILLIMUS (Gunther)Dipsas gracilUm<i GiJNTHEat, Biol. Centr. Amer. Kept. Bati'., p. 177, pi. 56, fig. B,1895.Himantodes gracillimus BotiLENGER, Cat. Snakes Brit. Mus., vol. 3, p. 87, 1896.Diagnosis.?Dorsal cross bands very numerous (74 in single speci-men examined), becoming broken and indistinct on posterior part ofbody; ventrals 244 to 253 (3 specimens).Range.?The west coast of Mexico, presumably from Nayarit tocentral Guerrero. The only definite records are from Tres MariasIslands (Boulenger loc. cit.) and Acapulco, Guerrero (M.C.Z. No. 823)
.
Specimens examined.?One.Bemurks.?The apparent existence in the area between Guerreroand Nayarit of three relatively closely related species of hnantodes isextraordinary and invites further attention as other specimens becomeavailable. At present, however, all three species seem very wellfounded. IMANTODES LATISTRATUS (Cope)Plate 37, Figure 2Dipsas gemmistrata latistrata Cope, U. S. Nat. Mus. Bull. 32, p. 68, 1887 (Guada-lajara and Valley of Toluca).Diagnosis.?Vertebral scales not enlarged; dorsal bands 33 to 67,nearly or quite as wide laterally as dorsally, middorsally separatedfrom each other by narrow light areas of about one scale length, noneof dark bands broken laterally as in splendidus., but the bands becom-ing indistinct and sometimes indistinguishable on posterior part ofbody in adults (in young, the bands remain distinct on all parts of thebody, but are not broken laterally) ; ventrals 223 to 233 (4 specimens),caudals 128 to 134.Range.?Nayarit southward to central Guerrero.Specimens examined.?Ten.RemarJcs.?This very distinct species appears to overlap the rangeof splendidus oliveri. Records available are from Miramar (U.S.N.M.Nos. 51481-51483) and Compostela (Univ. Eochester No. 5379), Nay-arit; ? Hda. Santa Gertrudis^"; Guadalajara (U.S.N.M. No. 24963),Jalisco; Hda. El Sabino (EHT-HMS No. ^?>^Q),Mi?honc/in; Huajint-lan (EHT-HMS No. 5205) and 6 miles northeast of Cuernavaca(EHT-HMS No. 5517), Morelos; Toluca (A.N.S.P. No. 11677) andMotaje,-^ Mexico; and Chilpancingo (M.C.Z. Nos. 33651-33652),Guerrero.
?? Boulenger, Cat. Snakes Brit. Mus., vol. 3, pp. 81-87, 1896.? Herreia. Cat. Rept. Batr. Mus. Nac. Mex., ed. 2, p. 30, 1904.
388 PROCEEDESTGS OF THE NATIONAL IMUSEUM vol. 92IMANTODES SPLENDIDUS SPLENDIDUS (Gunther)Dipsas splendida Gunther, Biol. Centr. Amer. Rept., p. 176, pi. 56, fig. A., 1S&5(Yucatan).Imantodes splcndidus Schmidt and Andrews, Field Miis. Nat. Hist., zool. ser.,vol. 20, pp. 170-177, 1936.Imantodes ffemmistratus Hartweg and Olivee, Misc. Publ. Univ. Mich. Miis.Zool., No. 47, p. 24 (part), 1940.Diagnosis.?^Vertebrals not or scarcely enlarged, none or very fewapproaching twice the width of adjacent scales ; usually 3 labials enterorbit ; all body bands distinct, about 34 to 39, most of them much nar-rower laterally than clorsally, only those on posterior part of bodybroken laterally ; ventrals 198 to 201.Range.?Northern part of the Yucatan peninsula./Specimens exam hied.?Two.Remarks.?With the restriction of the name gemmistratus to aspecies with larger vertebral scales, splendidus becomes available forthe Mexican species previously called gemmistratus. There are atleast 3 races, and probably more.IMANTODES SPLENDIDUS LUCIODORSUS OliverImantodes luciodorsus Ousee, Copeia, 1942, No. 1, pp. 1-2.Diagnosis.?Like s. splendidus, except: Dorsal body spots 43 to 52,average 46, broken laterally on posterior half of body; ventral 205 to225, average 219 (14 specimens).Range.-?Coastal or lowland regions on Atlantic slopes from centralVeracruz through Campeche and northern Chiapas into Guatemala,avoiding the northern half of the Yucatan peninsula.JSpeci.7ne7hs exaniin ed.?Twelve.Remarks.?With the definition of this race closely related tosplendidus., and the restriction of the name genfimistratus to a speciesof Imantodes with distinctly enlarged vertebral scales, a new name forthe Pacific coast race, closely related to luciodorsus., is necessai'y.IMANTODES SPLENDIDUS OUVERI, new subspeciesDipsas gemmistrata Ferraei-Perez, Proc. U. S. Nat. Mus., vol. 9, p. 185, 1886(Chiapas).?Cope. U. S. Nat. Mus. Bull. 3?, p. 68, 1887 (Chiapas).Himantodes gemmistratus Sumichrast, Bull. Soc. Zool. France, vol. 5, p. 184,18S0 (Tapana [Tapanatepec], Oaxaca).
?
Mocquaed, Miss. Sci. Mex., Rept.,livr. 16, pp. 917-918, pi. 74, fig. 4, 1908 (Isthmus of Tehuantepec).? ? Werner,Mitt. Nat. Hist. Mus. Hamburg, vol. 26, pp. 22&-230, 1909 (Hacienda deIxtapa, Nayarit).Imantodes gemmistratus Oliver, Occ. Pap. Mus. Zool. Univ. Mich., No. 360, pp.23-24, 1937 (Hacienda Albarradita, Colima) ; Hartweg and Oliver, Misc.Publ. Mus. Zool. Univ. Mich., No. 47, p. 24, 1940 (La Mixtequilla, Oaxaca),Holotype.?M.C.Z. No. 27800, Tapanatepec, Oaxaca.Paratypes.?Fiit^QW, including U.S.N.M. Nos. 12089, 32171, Mexico;Nos. 12443, 30164-30166, 30178-30179, Tehuantepec, Oaxaca; Nos.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 38930386-30388, Jiichitaii, Oaxaca; No. 110528, Tonala, Chiapas; M.C.Z.Nos. 27799, 27801, Tapanatepec, Oaxaca ; and Univ. Mich. Mus. Zool.No. 82603, La Mixtequilla, Oaxaca.Diagnosis.?Like splendidus splendidus and s. luciodorsus, except:Ventrals 222 to 234, average 229 (16 specimens) ; dorsal body bands47 to 67, average 57, broken laterally on all except about anterior thirdof body.Description of holotype.?Head twice as wide as neck; portion ofrostral visible from above narrow, but little shorter than distance be-tween anterior tips of nasals; internasals a third size of prefrontals;latter narrowly in contact with supraocular; frontal pentagonal,anterior edge nearly straight, posterior angle slightly less than aright angle; length of frontal (4 mm.) a third greater than its dis-tance from snout (3 mm.), a little longer than median suture be-tween parietals (3.5 mm.), about two-thirds maximum length ofparietals (5.5 mm.), and about a third greater than its maximumwidth (2.7 mm.); nasal completely divided, nares largely in an-terior portion, which is much higher and a little larger than posteriorpart; loreal nearly square, about as high as long; a single largepreocular ; postoculars 2-3, upper much the largest ; temporals 1-2-3
;
supralabials 8-8, sixth largest, third, fourth, and fifth entering orbit;infralabials 10-10, five in contact with anterior chinshields, 2 (fifthand sixth) with posterior; chinshields nearly subequal in size, pos-terior a little shorter than anterior and separated from each othermedially by a series of small scales ; 3 small scales between chinshieldsand first enlarged ventral.Dorsal scales in 17-17-15 rows, smooth, with 1 or 2 apical pitsjvertebral scales somewhat wider than others ; ventrals 222 ; subcaudals127; anal divided; total length 774 mm., tail 230 mm., female.Color.?Ground color light brown, stippled; a series of 58 crossbands on body extending to edges of ventrals, and covering about1^4 to 21/2 scale lengths middorsally; spaces between anterior bandsa little less than length of bands middorsally, but increasing in sizeposteriorly, the posterior bands separated by spaces which may be asmuch as II/2 times as long as the bands themselves ; a very irregular,broken, narrow, dark streak on middorsal line between bands; bandsa little narrower on sides than middorsally; anterior 20 bands un-broken laterally, but the twenty-first and all posterior bands arebroken laterally on the left side (at seventy-ninth ventral), while onthe right side the twenty-fourth band is the most anterior brokenband (at ninety-first ventral) ; the lateral break in the bands becomesmuch greater posteriorly so that there remains on the posterior partof the body a median series of large spots, extending laterally to thefifth scale row, separated from a lateral series of smaller, poorly de-fined spots involving the two edges of the ventrals and the first and
390 PROCEEDINGS OF THE NATIONAL MUSEUMsecond scale rows; the dorsal spots coincide in position with thelateral spots throughout the body.Dorsal spots on tail about 38, becoming smaller and less distinctdistally; large, dark brown stipple marks on belly and subcaudal sur-face, except on neck, chin and gular region, which are immaculate.Table 2.
?
Variation in Imautodes s. oliveri
Number
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 391KEY TO MEXICAN IMANTODES
1. Vertebral scales greatly enlarged, three to four times as wide as adjacentdorsals ; bands symmetrical, usually not broken on sides of body posteriorly.cenchoa leucomelasVertebral scales not enlarged or at most about twice as wide as adjacentscales 22. Ventrals 240 or more 3Ventrals 236 or fewer 43. Posterior dark bands on body broken laterally, poorly defined ; spaces betweenanterior cross bars less than one scale length gracillimusBands distinct and symmetrical over all of body, usually none broken on sidesof body, all involving ends of ventrals or subcaudals ; spaces between cross-bands half the length of the bands, or greater tenuissimus4. Vertebrals about twice as wide as adjacent scales gemm^istratusVertebrals not or but slightly enlarged, not twice as wide as adjacentdorsals 55. Bands with nearly straight anterior and posterior edges, very little if anynarrower laterally than dorsally; bands not broken on sides of body, butbecoming much less distinct toward tail, sometimes indistinguishable;ventrals 223 to 233 latistratusBands narrower on sides of body than on middorsum, at least the posteriorones broken laterally, and all very distinct and not becoming notablyfainter toward tail 66. Body bands 34 to 39, only those on extreme posterior part of body brokenlaterally; ventrals 198 to 201 splendidus splendidusBody bauds 43 or more, broken laterally on at least posterior half of body,sometimes on posterior two-thirds ; ventrals 205 or more 77. Body bands 47 to 67, average 57, broken laterally on all except anterior thirdof body; ventrals 222 to 235, average 229 splendidus oliveriBody bands 43 to 52, average 46, broken laterally only on posterior half ofbody ; ventrals 205 to 225, average 219 splendidus luciodorsus8.?TWO NEW SNAKES OF THE GENUS CLELIAAmong the very important discoveries of Dyfrig McH. Forbes ofPotrero Viejo, Veracruz, is a rare specimen, generously presented tome in 1939, belonging to the pethola group of Clelia. This is the thirdknown from Mexico, and apparently the first in American collections.Its characters do not agree with those of Central American specimens,and no name for it is available. It is a pleasure to associate with itthe name of Dr. Joseph R. Bailey, in whose hands the genus is in theprocess of intensive study.CLELIA BAILEYI, new speciesEolotype.?V.S.l^M. No. 111261, female, from Potrero Viejo, Vera-cruz, collected by Dyfrig McH. Forbes.Diagnosis.?Subcaudals divided; supralabials eight; preocular incontact with frontal; ventrals 191, caudals 89 in type; 23 crossbandson body, 13 on tail, in type; body bands covering four to five scale
392 PROCEEDINGS OF THE NATIONAL MUSEUM vol.82lengths, separated from each other by red spaces of nearly equal size
;
all except anterior 4 or 5 red interspaces with most of the scales spotted(black-tipped or bases black) ; first black band (nape) covering about8 scale lengths ; yellow nuchal collar covering three scale lengths mid-dorsally, laterally extending only to posterior portion of seventhsupralabials ; remainder of sides and top of head black.Description of holotype.?Rostral broader than high; length of j>or-tion visible from above three-fourths length of internasal suture ; inter-nasal two-fifths size of prefrontals, quadrangular; prefrontals ex-t/Cnding well onto sides of head, the greatest extension a point betweenloreal and preocular; frontal pentagonal, the lateral and posterolateralsides meeting in a broad curve; length of frontal (5 mm.) slightlyless than its distance from tip of snout (5.5 mm.) nearly as great asgreatest length of parietal (6 mm.) ; nasal large, completely divided,naris large and centrally placed, anterior section larger than posteriorand wedged between first labial and rostral; loreal twice as long asbroad; preocular single, nearly as large as supraocular, in contactwith frontal and with one (two) labials, much narrower below thanabove; two postoculars, lower smaller; temjiorals 2-2 (2-3), the sixthlabial narrowly in contact with upper primary temporal ; eight supra-labials, sixth and seventh subequal and larger than others, fourth andfifth (third also, narrowly, on one side) entering orbit; eye small,its diameter (2.8 mm.) about half its distance from tip of snout (5.8mm.) ; ten infralabials, sixth largest, five in contact with anterior chin-shields; latter a little larger than posterior, which are narrowly sepa-rated from each other.Dorsal scales in 19-19-17 rows, smooth, with 2 large apical pits;ventrals 191 ; anal single ; caudals 89. Total length 700 mm., tail 173mm.Maxilla with 15 teeth, the 2 posterior somewhat enlarged, and witha deep groove on anterior face; other teeth ungrooved; the anteriorones strongly recurved and a little larger than posterior teeth (ofungrooved series).Color.?Top and sides of head uniform black, this color extendinglaterally to posterior border of seventh supralabial, medially a littlebeyond end of parietals ; this bordered by a yellow nuchal collar witha straight posterior border, covering about 3 scale lengths medially;follows a series of 23 crossbands on body, 13 on tail; bands shorteron posterior part of body than on anterior, all a little shorter laterallythan medially; first band covering about 8 scale lengths, others 6 to31/^ scale lengths, decreasing posteriorly ; these bands involving endsof ventrals ; red spaces between black bands nearly equal size of lat-ter; anterior red space without black (a few tiny flecks) ; succeeding3 species with numerous, small flecks; succeeding 2 with the tip of
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 393
?ach scale black, and some larger dark marks medially; remainder ofred interspaces with numerous black spots, smaller than a scale, at tipsor bases of scales.Infralabial and lateral gular region with poorly defined, smallblack spots and fine black stippling; belly cream, unmarked save atends of ventrals (black spotted, bands also encroaching) ; ventralsurface of tail largely black, with irregular light areas and edges ofcaudals light, giving a mottled (not at all banded) appearance.Remarks.?The present species' nearest relative appears to be thatdescribed by Cope"- as Oxyrhopus doliatus semicinctus, the type ofwhich is U.S.N.M. No, 28900, from Sipurio, Costa Rica. Conspecificwith this are 4 others in the National Museum from Honduras (Se-govia River, No. 24533), Nicaragua (Escondido River, No. 19744) andPanama (Cana, No. 50111; Ancon, C. Z., No. 65867). These speci-mens show the following differences from haileyi: (1) ventrals morenumerous, 200 to 209 (191, haileyi)., caudals more numerous, 92 to111 (89, haileyi) ; (2) very little or no black spotting on light scales(a great deal in haileyi) ; (3) ventral surface of tail banded, as dorsalsurface (mottled in haileyi) ; (4) interspaces between black bandsyellow (posteriorly red?) (aU red in haileyi, except yellow nuchalcollar) ; (5) nuchal collar usually (not in 1) involving posterior endsof parietals, nearly to eye laterally, covering 5 to 8 scale lengths onnape (not involving parietals, not reaching primary temporals lat-erally, covering 3 scale lengths on nape) ; (6) first spot on bodylonger, covering 10 to 16 scale lengths (8 scale lengths in haileyi).So far as I am aware, the only other specimens of this group knownfrom Mexico are two recorded by Boulenger -^ from "Mexico" andAtoyac, Guerrero. The latter very likely is different from haileyi,not only because it has more numerous ventrals (204) but primarilybecause it is from an entirely different faunal province.This species is not pethola Linnaeus, since this name {-fide Boulen-ger, op. cit., p. 102) refers to South American specimens with 50 to75 crossbands; the type of haileyi moreover does not fit the descrip-tions given by Linnaeus of the 9 varieties of pethola. The patterndescription of petalarius Linnaeus does fit the characters of haileyi,however, in the few details mentioned; I do not believe the namerefers to the Mexican foiTn, however, since the ventrals and caudalsare perhaps too numerous (212, 102, respectively), and the type lo-cality is given as "South America" and "India" (latter by error).Oxyrhoipus doliatus aequifasciata Werner from Coban, Guatemala,does not seem to be the same. It has 207 ventrals and 78 subcaudals,the underside uniform yellow, the light spaces between the dorsalbands not dark-spotted.
=^ U. S. Nat. Mus. Bull. 32, p. 76, 1S87." Cat. Snakes Brit. Mus., vol. 3, p. 103, 1896.
394 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 92Remarks.?^The only two specimens of Clelia clelia examined fromthe Pacific slopes of Mexico north of the Isthmus of Tehuantepec aremarkedly different from other Mexican specimens in lacking darkspots at the tips of the dorsal scales. They are named
:
CLELIA CLELIA IMMACULATA, new subspecieaHolotype.?V.^.l^M. No. 24966, from "Guadalajara," Jalisco, col-lected by J. J. Major.Pamtype.?EWY-RMS No. 4568 from Paso del Rio, Colima.Diagnosis.?Like Clelia clelia, except no dark spots present on tipsof dorsal scales.Description of holotype.?Supralabials 7-7; infralabials 8-8; pos-terior section of nasal elongate; a small loreal; 1 preocular; 2 post-oculars; temporals, 2-3; dorsals 19-19-17 rows; ventrals 206; analentire; subcaudals 78. Total length 595 mm.; tail 118 mm.; male.Top of head dark brown, about to middle of parietals, the posterioredge irregular; nape and posterior part of head white; an area onnape, beginning three scale lengths behind parietals, dark brown, thecolor extending posteriorly about eight scale lengths, then becomingdiffuse and disappearing; remainder of dorsal surface light, cream(red in life), without any dark marks whatever; ventral surfaces ofhead, body and tail white.Variation.?The only other specimen seen from the Pacific slopesof Mexico north of the Isthmus of Tehuantepec is a specimen I col-lected at Paso del Rio, Colima. It also lacks dark spots at the tipsof the scales. "Ventrals, 211; caudals, 86; upper labials, 7-7; lowerlabials, 8-8; preoculars, 1-1; postoculars, 2-2; temporals, 2-3; analsingle; scale formula 24, 21, 17, 17, 17; preocular separated fromfrontal; parietals shorter than their distance to end of snout; frontalabout equal to its distance to end of snout. Eye (3 mm.) shorterthan distance between eye and nostril (3.6 mm.) ; loreal higher thanlong; length 675 mm.; tail, 147 mm.; head width, 12 mm.; length,19.6 mm. Head dark violet to violet-black; cream collar behindparietals 4 scale-rows wide; large blackish violet band behindcollar." 2*Comparisons.?Nine specimens of Clelia clelia clelia from Chiapas,,the Isthmus of Tehuantepec, and Veracruz all have the dorsal scalesblack-tipped. The black tips are visible even in adult specimens.9.?ADDITIONAL NOTES ON CONOPHISIn a recent paper on Con,ophis^^ I overlooked the notice byDunn -^ of a new species, C. never?7Uinni, from Costa Rica. The
?< Taylor and Smith, Kans. Univ. Sci. Bull., vol. 25, p. 258, 1938.^ Jour. Washington Acad. Sci., vol. 31, No. 3, pp. 117-124, 1941.*" Copeia, 1937, No. 4, p. 214, 1938.
MEXICAN HERPETOLOGICAL MISCELLANY?SMITH 395latter is a species I have not seen, and it is not represented in thecollection of the United States National Museum. It possibly be-longs in the group with pulcher and Uneatus, since it has eightsupralabials and presumably the first scale row pigmented. It seemsclosest to pulcher, of which it could conceivably be a race. It cer-tainly is a species different from Uneatus, which also occurs in CostaRica.The Central American subspecies of Uneatus, which I namedsimilis, cannot stand, since it is a homonym of Gonophis pulcherslmilis Bocourt.-^ I suggest the name dunnl for the southern race;its type (that of similw Smith) is U.S.N.M. No. 799G3, from Ma-nagua, Nicaragua.Unfortunately Bocourt's pulcher similis was described without ci-tation of the type locality, and many details of its pattern are notmentioned. However, in several respects it agrees with G. pulcherplagosus, which I described on the basis of a single specimen fromTonala, Chiapas (U.S.N.M. No. 109707). Akhough absolute cer-tainty of the identity of Bocourt's specimen can be obtained only byreexamination of the type, it appears probable that plagosus Smithand similis Bocourt are identical.The species of the genus accordingly stand as follows
:
GonopMs vittatus vittatus PetersG&nophis vittatus viduu^ CopeConophis pulcher pulcher CopeGonophis pulcher similis BocourtGonophis nevermanni DunnGonophis Uneatus Uneatus (Dumeril and Bibron)Gonophis Uneatus concolor CopeGo7iophis Uneatus dunni Smith.
?? Miss. Sci. Mex., Rept., livr. 10, 1886, pp. 643-4.