632 PROCEEDINGS OF UNITED STA.TES NATIONAL MUSEUM.covered with minute, cup-sbaped depressions, into which open the per-forations of the tube. Distance from tentacle to eye but twice lengthof tentacle, which equals one-fourth diameter of orbit.Caudal fin equal to length of caudal peduncle. Dorsal large, the baseequaling three-sevenths height of fin.Body without fleshy slips or folds.Head 3^ in length; eye about one-fourth head. D. or 10; A. 10.Color in spirits : Dark brown above, white below ; entire upi^er partsincluding caudal fin, covered with round, white spots, most numerouson caudal peduncle, the largest much less than half pupil; a round blackarea surrounding base of pectorals, bounded by a white line; severalparallel longitudinal black streaks below the pectorals ; orbit with twoconcentric white rings.Known from six specimens collected by Professor Bradley at Panama.The type is numbered 29079 on the register of the National Museum.The following species are here for the first time recorded from Panama:1. Ginglymostoma cirratum.2. Urolophus halleri.3. Syrrhina exasperata.4. Stolephorus miarchus.5. Ophisurus xysturus.G. Oi)hichthys zophochir.7. Sidera verrilli sp. nov.8. Serranus calopteryx.9. Pomadasys brevipinnis.10. Gobiesox adustus.11. Emblemaria nivipes sp. nov.12. Salarias rubropunctatus.13. Dactyloscopus sp. nov. (?)11. Tetrodon angusticeps.15. Arothron erethizon sp. nov.Indiana University, December 1, 1882. jriTMPIiVG SEEDS AND GAIiliS.*By CHARLES V. RII^EIT.Having recently received some fresh specimens of so-called "MexicanJumping Seeds," or "Devil's Beans," as they are popularly called, I takeoccasion while yet they are active to exhibit them to the society. Itwill be noticed that these seeds are somewhat triangular, or of the shapeof convolvulus seeds, there being two flat sides meeting at an obtuseangle, and a convex one, which has a median carina. They not only *Rcacl before the Biological Society of Wasliingtou November 24, 1882. PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 633 roll from oue side to another, but actually move by jerks and jumps,and will, when very active, jump at least a liue from any object tliey Carpocapsa SALTiTANS : rt, larva; ft, pupa; c, imago?enlarged, bair-lines showinguat. size; d, front wing of a pale var. ; e, seed, uat. size, with empty pupa skin; /,do. showing hole of exit.?(After Ri'ey.)may be resting on. The actual jumping power has been doubted bysome writers, but I have often witnessed it. To the uninitiated thesemovements of a hard seed seem little less than miraculous. They areinduced by a plump, whitish, lepidopterous larva which occupies aboutone-fifth of the interior, the occnpied seed being, in fact, but ahollow shell,with an inner lining of silk which the larva has spun. The larva looksvery much like the common api)le-worm {Carpocapsa pomonella), and be-longs, in fact, to the same genus. It resembles it further in remaining fora long time in the full-grown larva state before transforming, so that theseeds will keep up their motion throughout most of the winter months.When about to transform, which is usually in the months of Januaryand February, it cuts a neat, circular door in the convex side of its house,strengthens the same with silk, spins a loose tube of silk within theseed, and therein transforms to the pupa state. The moth soon after-ward pushes its way out from the little door prepared for it.The moth was first described in 1857 as Carpocapsa saltitans by Prof.J. O. Westwood,* and afterward as Carpocapsa dehaisiana by Mous. H.Lucas, tIn regard to the plant on which these seeds occur there is much yetto learn, and I quote what Mr. G. W. Barnes, president of the SanDiego Society of Natural History, wrote me in 1874 concerning it, in thehope that some of the botanists present may recognize it:"Arrow-weed {Yerba cleflecha).?This is the name the shrub bearsthat produces the triangular seeds that during six or eight months havea continual jumping movement. The shrub is small, from 4 to G feet inheight, branchy, and in the months of June and July yields the seeds,a pod containing three to five seeds. These seeds have each a littleworm inside. The leaf of the plant is very similar to that of the ga- *Proc. Ashmolean Soc. of Oxford, 1857, t, 3, pp. 137-8; see also Trans. Lond. Ent.Soc., ser. 2, 1858, t. IV, p. 27, 'and Gard. Chron. 1859, Nov. 12, p. 909.+ "Note sur les grains d'uue Euphorbiac^e de Mexique sautant an dessus du sol parles vibrations d'nne larve de I'ordre des lepidopteres vivant en dedans."? (Ann. Soc.Ent de France, ser. 3, t. 6, Bull. p. 10, p. 33, p. 41, p. 44, 1859; t. 7, p. 561-566.) 634 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM.rambullo, the ouly difierence being in the size, this being a little larger.It is half an inch in length and a quarter of an inch in width, a littlemore or less. The bark of the shrub is ash-colored, and the leaf is per-fectly green during all the seasons. By merely stirring coffee, or anydrink, with a small branch of it, it acts as an active cathartic. Takenin large doses it is an active poison, speedily causing death unless coun-teracted by an antidote."lu a recent letter he states that he is informed that the region ofMamos, in Sonora, is the only place where the plant grows ; that thetree is about four feet high and is a species of laurel, with the leaves ofa dark varnished green. " It bears the seeds only once in two years.The tree is called Brincador (jumper), and the seeds are called Brinca-deros. The seeds are more quiet in fair weather, and lively on the ap-proach of a storm."Professor Westwood mentions the fact that the jjlant is known bythe Mexicans as ^^ Colliguaja ;" and Prof. E. P. Cox, formerly Stategeologist of Indiana, now living on the Pacific coast, informs me thatthe shrub has a wood something like hazel or whahoo; that the leaf islike a broad and short willow leaf. He confirms the statement as toits poisonous character; that a stick of the shrub, when used by thenatives to stir their "penola" (ground corn-meal, parched), purges, andthat the shrub is used to poison arrowheads. The plant is undoubtedlyEuphorbiaceous. *The peculiarity about this insect is that it is the only one of its order,so far as we know, which possesses this habit, and it is not easy to con-ceive of what benefit this habit can be other than the possible pro-tection aftbrded by working the seed, after it falls to the ground, intosheltered situations.The true explanation of the movements of the larva by which theseed is made to jump was first given by me in the Transactions of theSaint Louis Academy of Science for December 6, 1875 (Vol. Ill, p.cxci).The jumping power exhibited in this " seed" is, however, trifling com,pared with that possessed in a little gall which I also exhibit. Thisgall, about the size of a mustard seed, and looking very much like aminiature acorn, is found in large numbers on the under side of theleaves of various oaks of the White Oak group, and has been reportedfrom Ohio, Indiana, Missouri, and California. It falls from a cavity onthe under side of the leaves, very much as an acorn falls from its cup,and is sometimes so abundant that the ground beneath an infested treeis literally covered. It is produced by a little black Cynips, whichwas described as Cynips saltatorius by Mr. Henry Edwards. Thebounding motion is doubtless caused by the larva which lies curvedwithin the gall, and very much on the same principle that the commoncheese-skii)per {FiopMla casei) is known to spring or skip. Dr. W.H. Mussey, of Cincinnati, in a communication to the Natural HistorySociety of that city, December, 1875, states, in fact, that such is the PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 635case ; though members of the California Academy who have writtenon the subject assert that the motion is made by the pupa, which Ithink very improbable. At all events the bounding motion is great,as the little gall may be thrown 2 or 3 inches from the earth ; andthere are few things more curious than to witness, as I have done,a large number of these tiny galls in constant motion under a tree.They cause a noise upon the fallen leaves that may be likened to thepattering of rain. WOTE ON CliUSTER FL.IES.By W. H. DALI..Having heard several years ago of a fly which was a great nuisancein the country houses near Geneva, N. Y., among members of my wife'sfamily living there, I requested information and specimens when itshould be convenient. Some time since a relative visited Geneva, andon his return brought me some of these flies alive in a bottle coveredwith gauze, which were exhibited at the last meeting of the BiologicalSociety and turned over to Prof. C. V. Eiley for identification. Sincethen a letter has been received, from which I make the following extracts : "It is probably thirty years since the flies appeared in our neighbor-hood. I remember little about it except that they were at once a terrorto all neat housekeepers, and from their peculiar habits a constant sur-prise. People soon learned to lo?k for them everywhere; in beds, inpillow slips, nuder table covers, behind pictures, in wardrobes nestledin bonnets and hats, under the edge of carpets, and in all i)ossible andimpossible places. A window casing st?lidly nailed on will, when re-moved, show a solid line of them from top to bottom; they are uncanny.They like new houses, but are often found swarming in old unusedbuildings and go regularly to church, or perhaps ouly a few good onesabide in sanctuaries ; any way they are there. Best of all tbey like aclean dark chamber seldom used, and if not disturbed form in largeclusters about the ceilings. With tliem are usually found a number ofpurplish black hornets and some ladybugs ( Coccinella). They are verycold and feel in the hand like small bits of ice. They are very oily; ifcrushed, leave on the floor a great grease-spot. I hardly think theybreed in the houses, but do not know. About the 1st of April or assoon as the sun shines warm in the early spring they come out in thegrass and fly up to the sunny side of the houses. Some possibly creepin open windows, or if the house is closed and sealed they have a facultyof going through any crack. They remain until some time in May,then disappear, and no more are seen until about September, when theycome and remain as long as they are allowed to. They are very strong.A powder that suffocates common house flies has very little effect onthem, and we attack them with ammonia and drown them with boilingwater; even then are not sure they are 'kilt entirely.' Very few arefound in the towns or villages ; they live in country places altogether.