A NEW SPECIES OF BULLFROG FROM FLORIDA ANDTHE GULF COAST. By Leonhakd Stejneger,Curator, Division of Reptiles and BatracMans.A few years ago Mr. RolxTt Ridgwa}', returning from a collectingtrip in southern Florida, assured me that the bullfrog there had sucha peculiar voice that he could scarcely believe it to be the samespecies as the one found elsewhere in the United States. Moreover,it was so excessively shy and wary that he had failed to secure anyspecimens. Requests were sent to various correspondents to catchand forwai'd specimens, but ih&y were equally unsuccessful.During the summer of 1900 Mr. Gerrit S. Miller, jr., obtained aseries of live bullfrogs from New Hauipshire and another from Ba}^St. Louis, Mississippi, all of which he kindl}" donated to the Museum.It w^as at once apparent that there was a great difference between thetwo lots, the northern ones being much greener and lighter, thesouthern specimens being brown and dark, with very little green.The large series of alcoholic bullfrogs in the collection of the U. S.National Museum failed to throw an}" light upon this question. Thedifference of the proportions of the toes was also noted, but as speci-mens from New Orleans did not correspond in this respect the impor-tance of that character was not realized at the time, and the questionof the specific or subspecitic distinction of the Bay St. Louis speci-mens was left in abej^ance pending the accumulation of additionalmaterial.This additional material was obtained recently when Dr. E. A.Mearns, U. S. A., sent a lot of bullfrogs from Kissimmee, Florida,with the statement that their voice was entirely different from that ofthe northern bullfrogs, resembling the grunt of a herd of pigs. Healso reported that the}^ were very difficult to catch.Here was clearly the Florida bullfrog mentioned by Mr. Ridgway,and a direct comparison with the Bay St. Louis specimens proved theidentity of the Florida and Mississippi form. The whole series ofthe museum was then carefully reviewed, with the result of finding twoadditional specimens of the new species from Pensacola.Proceedings U. S. National Museum, Vol. XXIV?No. 1252. 211 212 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. XXIV.RANA GRYLIO, new species.Dia(/nos/s.?Similar to Rana catesbeiana^ but with the fourth toemuch shorter in proportion, the third toe, measured from the innermetatarsal tubercle, being more than three and one-half times thedifference between the third and fourth toes.Type.?C?it. No. 27443, U.S.N.M.; Bay St. Louis, Mississippi.Ilahitat.?Florida and Gulf coast west to Mississippi.Remarks.?The most obvious difference between this species andthe ordinary bullfrog- {Rana catesbelana) is the great length of the toes,except the fourth, the latter consequently projecting much less beyondthe others than in R. catesheiana., in which the third toe, measuredfrom the inner metatarsal tubercle, is considerablj^ less than three andone-half times the difference between it and the fourth toe. In fact,this difference is seldom more than one-fourth in the new species andseldom less than one-third in R. catesheiana.In order to ascertain exactly the proportions of the first four toesin both species large series of both species were measured, viz: 13 ofR. grylio and 50 of R. catesbekma., the measurements, as well as theirequivalents expressed in percentages of the fourth toe, being given atthe end of this article. In order to get as stable a starting point aspossible for these measurements the anterior edge of the inner meta-tarsal tubercle was chosen and the length of the toe in this case conse-quently means the distance from this point to the tip of the toe inquestion.The proportions obtained in this way may lie expressed as follows: A NEW BULLFROG?STEJNEGER. 213 'riie third toe, .siiiiihirly niotisured, 41 niiii. (varviniL?- l)otweoii -tOnull, and 4li iiini.).Conversely, a Rami aitcshcldud of exactly the .same size (fourtli toe50 mm.), should normally have ? The tirst toe, similarly measured, 15 mm. (varying ])etween 13.5mm. and 1(),5 mm.).The second toe. similarly measured, ^4.5 mm. (varying between 23.5mm. and 25.5 mm.).The third toe, similarly measured, 36.5 mm. (varying between 35mm. and 38 mm.).These average jjroportions are shown diagram maticallj" in the accom-panying figure, in which the dotted outline represents an average B.yn/Iio and the solid lines an average liidia catesbelana.These ditt'erences in the relative length of the toes, being capable ofthe most concise definition, have been utilizedprimarily for the characterization of the newspecies, but there are numerous other fea-tures which prove it to be very distinct fromR. catesheiana. Thus for instance, the snoutis much shorter and less high, so that thenostrils appear to be nearer the tip of themouth; the head is also narrower behind; the\omerine teeth are very close together, withhardly any space l^etween the two patches;the tongue is much broader and thinner, withremarkal)ly long and thin "horns," which arevery far apart; the color is apparently muchdarker brown, though there may be li. cates-hrlana nearly as dark. Add to this the differ-ence in voice and we have clearly one of the mostdistinct species of frog in the United States.The general habitus of R. grylio is that ofR. catesheiana^ including the large tympanum and the absence of adorso-lateral glandular fold. It is also a large frog, though whether itreaches the extreme size of R. catesbelana ma}^ be doubted.The range of the new species is as yet known but fragmentarily, aswe have specimens only from southern Florida, Pensacola, Florida,and Bay St. Louis, Mississippi. The habitat of R. grylio is thusparth^ occupied by R. (Xitesheicma, which certainl}^ occurs in northernFlorida and on the Gulf coast reaches New Orleans. The overlappingof the two forms afi'ords additional evidence of their specific distinct-ness, if such were needed. It ma}^ be mentioned in this connectionthat the most southern specimens of R. catesbeiana do not show theslightest tendency of a variation toward R. grylio^ as is clear!}- provenby the measurements in the appended tables. DiAGEAMMATIC OUTLINES OFHIND FEET OF R. catcsbeiaiiaAND R. grylio, the formerIN SOLID LINES AND THE LAT-TER DOTTED. 214 PROCEEDINGS OF THE NATIONAL MUSEUM. vol,. XXIV. 3fr(isiirt'iii('iilx of li