DEPARTMENT OF MARINE BIOLOGY OF THE CARNEGIE INSTITUTION OF WASHINGTON ALFRED G. MAYER, DIRECTOR PAPERS FROM THE DEPARTMENT OF MARINE BIOLOGY OF THE CARNEGIE INSTITUTION OF WASHINGTON VOLUME XII WASHINGTON, D. C. PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON 1918 CARNEGIE INSTITUTION OF WASHINGTON PUBLICATION No. 252 PRESS OF GIBSON BROTHERS, INC. WASHINGTON, D. C. CONTENTS PAGE. I. Some Amphibians and Reptiles from Porto Rico and the Virgin Islands. By HENRY W. FOWLER. 1 plate and 6 figures 1-15 Leptodactylus albilabris (Gvinther) 3 Eleutherodactylus auriculatus (Cope) 4 Eleutherodactylus lentus (Cope) 6 Hemiclactylus rnabouia (Moreau de Jonnes) 7 Sphaerodactylus grandisquamis Stejneger 7 Mabuya sloanii (Daudin) 7 Anolis cuvieri Merrem 7 Anolis mayeri sp. nov 8 Anolis gundlachii Peters 9 Anolis cristatellus Durneril and Bibron 10 Anolis stratulus Cope 11 Anolis krugi Peters 11 Anolis pulchellus Dume'ril and Bibron 12 Anolis poncensis Stejneger 12 Ameiva exul (Cope) 12 Ameiva wetmorei Stejneger 12 Amphisbsena cseca Cuvier 14 Epicrates inornatus (Reinhardt) 14 Leimadophis stahli Stejneger 14 Alsophis portoricensis Reinhardt andyLutken 14 Alsophis antillensis (Schlegel) • 14 Alsophis rufiventris Dume'ril and Bibron 15 Pseudemys palustris (Gmelin) 15 II. Fishes New to the Fauna of Porto Rico, with Descriptions of Eight Neiv Species. By CHARLES F. SILVESTER. 4 plates 17-24 Aphthalmichthys mayeri sp. nov 19 Myrophis longleii sp. nov 20 Myrichthys kecki sp. nov 20 Tylosurus notatus (Poey) 21 Corythroichthys ensenadae sp. nov 21 Amia conklini sp. nov 21 Apogonichthys stellatus Cope 22 Centropomus pectinatus Poey 22 Diplodus argenteus (Cuvier and Valenciennes) 22 Eques pulcher Steindachner 22 Microspathodon niveatus (Poey) 22 Microspathoden chrysurus (Cuvier and Valenciennes) 22 Microspathodon fowleri sp. nov 22 Bodianus rufus (Linmeus) 23 Iridio maculipinna (Miiller and Troschel) 23 Pseudomonacanthus amphioxys (Cope) 23 Sicydium antillarum Ogilvie-Grant 23 Gobiosonaa viridistriatum sp. nov 24 Gobiesox cerasinus Cope 24 Alticus macclurei sp. nov 24 III iv Contents. PAGE. III. Oral Gestation in the Gaff-Topsail Catfish, Felichthys felis. By E. W. GUDGER. 4 plates 25-52 Narrative 27 Historical Account 29 The Breeding Season 30 Distinction between Sexes 32 Method of Transfer of the Eggs 32 Sex of the Egg Carrier 33 Size of Mouth Cavity 34 Size of Eggs and Number Incubated 35 Size of Young and Number Carried 37 Food and Feeding 39 Feeding and Growth of Young while in Paternal Mouth 42 Length of Period of Gestation 43 The Purpose of Buccal Incubation in the Gaff -Topsail 46 The Origin of the Habit of Oral Gestation 46 Literature Cited 51 IV. .S/'%''" '"•" Barracuda; Its Morphology, Habits, and History. By E. W. GTJDGER. 7 plates, 5 text-figures 53-108 General Description 55 Color and Markings 56 Jaws and Teeth 59 Manner of Breathing 70 Internal Organs 70 Habits 74 How the Barracuda may be taken 77 Food and Feeding 79 Parasites 80 Use of the Barracuda as Food and Poisoning resulting therefrom 83 Size 93 The Barracuda Dangerous to Man 94 Habitat 99 Fossil Forms \ 100 Nomenclature .\ 101 Drawings or Figures of the Barracuda 104 Literature Cited 107 V. Botanical Ecology of the Dry Tortitgas. By H. H. M. BOWMAN. 6 plates, 7 text-figures 109-13X Geologic Formation Ill Climatic Conditions of the Tort ugas 1 1 : '. General Sketch of the Vegetation of the Tortugas 115 Distribution of Species among the Keys of the Group IIS Loggerhead Key 1 20 Bird Key 123 Garden Key 125 Bush and Long Keys 128 Sand or Hospital Key 130 Middle Key " 131 East Key 131 Special Ecology 132 Marine Ecology 137 Conclusion.. 138 Contents. v PACK. VI. The Origin of the Electric Organs in Astroscopus guttalus. By E. GRACE WHITE. 7 plates, 1 text-figure 139-172 Historical Survey of the Subject 142 Observations 150 Experimental 162 Discussion 163 Summary 166 Bibliography 168 Explanation of Plates 171 VII. Toxic Effects due to High Temperature. By ALFRED GOLDSBOROTJGH MAYER. . 173-178 Summary 178 Literature Cited 178 VIII. Nerve-Conduction in Diluted and in Concentrated Sea-Water. By ALFRED GOLDSBOROTJGH MAYER. 1 text-figure 179-183 IX. A Study of Respiration in Alcyonaria. By LEWIS R. GARY 185-191 Experimental 187 Discussion 189 Relation between Death Temperature and Rate of Metabolism 190 Literature Cited 191 X. Migration of Insects to Rebecca Shoal Light-station and the Tortugas Islands, with *[/<'cial reference to Mosquitoes and Flies. By S. C. BALL 193-212 Moans by which Insects may reach Rebecca Shoal Light-house 196 ( >pportunities for Breeding of Mosquitoes at Rebecca Shoal 198 Conditions at Rebecca Shoal Light-station just prior to June 26 199 Methods of obtaining Insects which visited the station 200 Insects taken at Rebecca Shoal between June 26 and July 18, 1917 201 Migration of the House-fly 207 Occurrence of other Diptera at Rebecca Shoal 210 Other Insects noted at Rebecca Shoal 210 Evidence that Odonata and Leptidoptera Migrate to Tortugas 210 Do Sarcophagidse breed at Tortugas? 211 Summary 212 Bibliography 212 XI. On Changes in the Sea and their Relation to Organisms. By J. F. McCLENDON. 8 figures 213-259 Methods 218 Currents 221 Records of all Stations and of a Tank of Sea-water 222 Relation of Oxygen Tension and Activity of the Nervous System to Metabolism in Cassiopea 235 On the Chemical Precipitation of CaCO3 in Sea-water 252 Bibliography 259 I. SOME AMPHIBIANS AND REPTILES FROM PORTO RICO AND THE VIRGIN ISLANDS. BY HENRY W. FOWLER, Of the Academy of Natural Sciences of Philadelphia. One plate, six figures. SOME AMPHIBIANS AND REPTILES FROM PORTO RICO AND THE VIRGIN ISLANDS. BY HENRY W. FOWLER. The animals comprising the above-named groups were largely obtained in Porto Rico during the early summer of 1915 by Mr. Charles F. Silvester, while on the staff of the expedition of the Carnegie Insti- tution of Washington. They form an interesting accession to the Museum of Princeton University. Another collection, obtained in the Virgin Islands in 1876 by Mr. A. D. Brown, has been studied in this connection and included in the present report. It is also the property of Princeton University. I am under obligations to Mr. Silvester for the opportunity to study all this material, as well as for various notes, information, etc., kindly placed at my disposal. A series of specimens has been reserved for the Museum of the Academy of Natural Sciences of Philadelphia. The herpetology of Porto Rico has been so ably discussed by Dr. Leonhard Stejneger, in his paper of that title, published in 1902, that the present account is intended as merely supplementary. AMPHIBIA. LEPTODACTYLID^:. Leptodactylus albilabris (Giinther). Figure 1. Ten from the junction of the Arecibo and Tanama Rivers, Porto Rico, on June 9, 1915. As considerable variation in color is noticeable, 5 examples have been selected as an illustration of the dorsal aspect. Two adults show a strongly contrasted color-pattern above, largely of longitudinal bands, of which the median or vertebral is whitish, edged on each side by a blackish band of variable width in places; these blackish bands largely alike. Some- times a whitish vertebral band gives off a branch on each side in the front interorbital region, which extends over eyelid. Blackish band on each side of snout, including nostril, to eye, then continued behind to embrace tym- panum to shoulder, present in all examples. Side of each lip dusky. Several blackish spots on groin and a few obsolete dusky markings along the side. Upper surface of hind limbs strongly contrasted with blackish crossbars. Other examples show lengthwise bands on back obscure, or as irregular blotches, and along the sides as broken small spots. They often form two large A-shaped obscure blotches on back before pelvis. Two dark lateral bands may unite in hind interorbital region to form a connecting dark bar. Some examples have the throat dusted with pale dusky, though in most the entire under surface is uniform whitish. One metamorphosing example and several tadpoles from Utuado; a small example from the Virgin Islands. The Porto Rican examples were obtained along small streams flowing into the Arecibo River about 5 kilometers south of Arecibo, where the lowlands merge into the hills. Active during day-time. 3 4 Papers from the Department of Marine Biology. Eleutherodactylus auriculatus (Cope). Figure 2. Collected 21 examples from the junction of the Arecibo and Tanama Rivers on June 9. This little amphibian is extremely variable in color. Some individuals are pale or very light, others dark, and some of all ages with a narrow whitish vertebral line. Four extremes of color-variation are represented in the accompanying figures of the dorsal surface. All the examples with the narrow whitish vertebral line (4 individuals) show it bifurcate and extending out on each side of the hind femoral surface toward the under surface of the knee. The larger of these examples are also FIG. 1. — Leptodactylus albilalris (Gtinther). A to E, variation in dorsal color — pattern of examples of all a»cs from Areoibo and Tanama River Junction. Amphibians and Reptiles from Porto Rico and the Virgin Islands. 5 irregularly spotted or blotched with whitish on the back, about the inter- orbital region, and at each heel. A few spots may also occur elsewhere on the upper surfaces of the tibial regions. The smaller examples have large obscure blotches of slightly darker shade than the general color of the upper surface, which tend to form obscure bars across the tibial and femoral regions. The lower body-surface is similarly variable, from slightly soiled brownish to more or less completely marbled with smutty, in which case there is also a median obscure pale line from the chin to the sternum, with line given off each side to the axillaries. Fio. 2. — Eleutherodactylus auriculatus (Cope). A to D, variation in dorsal color — pattern of adult examples from Arecibo and Tanama River Junction. 6 Papers from the Department of Marine Biology. Four pale examples are also quite variable. One has the front and hind surfaces of the femoral region with rather large dusky vermiculations strongly contrasted with whitish. In this individual there is also a dark-gray inter- orbital band. Another shows a number of whitish spots on the middle of the back with broad whitish band back from each eye to groin. The lower sur- face of the body is often uniform whitish. Large darker olive or brownish examples are more or less uniform in color, though one shows pale areas back from the eye to the groin. They usually have the right and front femoral regions and groin finely speckled brownish on pale ground-color. Most are more or less smutty below. In a number of small darker examples traces of the light areas back from the eye are seen in two cases and in one of these an accessory short inner line backward and parallel is also present, while the larger bands tend to converge at the middle of the back. All have the limbs above with more or less obso- lete dark cross-bars. A dark interorbital band is also usually in evidence. All examples have the dark line from the eye over the tympanum distinct. Eleutherodactylus lentus (Cope). Figure 3. Twenty-one examples from the Virgin Islands. These agree with Meer- warth's figure of Hylodes lentus,1 based on examples from St. Thomas. The FK;. 3. — Eleutherodactylus lentus (Cope). From Virgin Islands. A, dorsal surface. B, ventral surface. (', si-It- view of head. D, open mouth, showing tongue and roof of mouth. 'Mitth. Naturh. Mus. Hamburjr, xvin, 1901, p. 38, pi. 2, figs. 1, 2. Amphibians and Reptiles from Porto Rico and the Virgin Islands. 7 species appears to be less variable than many of this genus. Although Cope states that the body " is without granulations above, below, or upon the sides, " all my examples show them on the lower hind femoral region. As Meerwarth s figure lacks detail I have made the following note. In this species the color- pattern is largely constant. Sometimes the blotches or vermiculations are larger and fewer on the back anteriorly, or they may vary in being finer and more numerous, though the general appearance is similar. REPTILIA. GEKKONID^. Hemidactylus mabouia (Moreau de Jonnes). Two from the Virgin Islands. According to Dr. Barbour,1 it was probably introduced into the West Indies. Sphaerodactylus grandisquamis Stejneger. Three from the Arecibo Road, Porto Rico, about the 70-kilometer post, on June 14; these are similar to Dr. Stejneger's figure,2 except that the black scapular blotch has in each example two small pure white spots. Two speci- mens have the dark spots as more or less broken longitudinal bands, though in the remaining example a more speckled appearance is seen and the spots are smaller. Though only the above examples were secured, this gecko was probably abundant at this locality. They were found under stones and sticks in moist places. Their tails were very brittle and were easily broken off. Mabuya sloanii (Daudin). Two from the Virgin Islands. One shows at least two pairs of chin shields in contact. Color in alcohol largely grayish olive above, paler or whitish below. From each side of the fronto-nasal plate a brown band extends back, including the upper eyelid, and becomes obscure toward the groin, though on front of body in strong contrast, due largely to darker brownish marginal lines. Another band of similar color extends from the nostrils back, including eye, over ear and shoulder, and fades out on groin. Limbs above with obscure brownish spots. Snout to vent 76 mm., tail 66 mm. This species is now rare in Porto Rico, a circumstance believed to be due to the mongoose. IGUANID^J. Anolis cuvieri Merrem. One from Utuado, 405 mm. in total length. Color in life with iris brown- ish yellow, a ring of pure yellow next to the black pupil. Scales around eye yellow, barred with black. These colors seem to hold even when the animal changes color from black to green and yellow. In alcohol the color is largely uniform bluish above and paler below. Pale crescent below eye. Known to the natives as "the devil." They are afraid of this species and believe that it is deadly poisonous. Mr. Silvester informs me it is rare and occurs along water-courses. He saw another example along the main course of the Arecibo River at Utuado, but was unable to capture it. 'Mem. Mus. Comp. ZooL, XLIV, No. 2, 1914, p. 222. 2Rep. U. S. Nat. Mus., 1902, p. 605, fig. 52. 8 Papers from the Department of Marine Biology. Anolis mayeri sp. nov. Figure 4. Description. — Body long, rather robust, with slight dorso-nuchal fold. Head rather level on top, with broad median depression on pref rental region. Slight median depression on occipital region. Eyelids granular. Teeth mod- erately large, and become enlarged posteriorly or below eye, those in front of each jaw small. Dewlap moderate, covered with small imbricated scales. Interorbital slightly convex. Ear-opening circular, deep. Scales all smooth, rather large. Median or vertebral row of scales down back slightly enlarged, and progressing down sides scales become granular on latter. Very minute granules seen scattered about between scales of back, though not visible to the naked eye and not preventing their imbrication on one or more sides in most cases. Scales on belly larger than most scales on back, and small on under surface of head. Scales on head above rather large and smooth. Supraoculars large, flat, and separated from supraorbital semicircles by narrow, complete row of small scales. Supraorbital semicircles with one or two median plates in contact, and separated from occipital plate by one or two rows of scales. Occipital large, nearly equals ear-opening, and scales around variably large and small. Counting vertically, six rows of loreal scales present, and/ of more or less uniform size. Six or seven large infraorbitals, last ending well before hind edge of eye. Temporals moderate, and about 12 scales between hind eye-edge and ear. Limbs rather robust. Scales on upper sides of fore-limbs larger than those on middle of back, slightly keeled, and closely imbricated. Scales along lower front femoral region larger than those on belly, smooth and closely imbricated. Scales on upper surface of femur very small and imbricated. Post-anal plate moderate. Tail strongly compressed, largely with slight crest. Scales covering sides of tail flat, keeled, those in crest longest and enlarged, a keel forming front edge of each. Color largely uniform brownish in alcohol, paler below. Some very obscure markings on back, upper surfaces of limbs and tail. Slight pale streak from eye through ear and backward. Another from over shoulder and backward. Total length 188 mm. (caudal tip damaged) and length from snout tip to vent 67 mm. Type No. 3151, Princeton University, and 7 paratypes. Virgin Islands. These examples suggest in many ways that they may be the young of Anolis cuvieri. They agree partly in having the scales of the back in places slightly separated from one another by the interpolating of more or less scat- tered, inconspicuous, and irregular small or minute granules, though these in no way encircle the scales. In general the scales of the back present an imbricated appearance. I have for these reasons allowed the species to stand near Anolis cuvieri. Its differences are, however, slight, and as the young of Anolis cuvieri is unknown, I first thought it to be such, especially as all my examples are comparatively small. It would appear too great a latitude in variation to admit such characters as these specimens present as simply variations due to age. First is the contact of several pairs of the large scutes of the supraorbital semicircles, rarely separated by a single row of very small scales. Second, the occipital scale is distinct and usually large, though often separated by only one scale from the supraorbital semicircles. The labials are in contact with the infraorbital semicircle, a character constant in every specimen. In agreement with Anolis cuvieri are the rows of large scales between the rictus and the ear, which vary from 5 to 8. This character will Amphibians and Reptiles from Porto Rico and the Virgin Islands. 9 readily serve to distinguish it from the common Anolis cristatellus. The present new form presents a considerable resemblance to Anolis monensis Stejneger (known only from Mona Island) in the larger scales and rather uniform coloration, though it has small post-rictal scales like its near ally Anolis cristatellus. Named for Dr. Alfred G. Mayer, of the Carnegie Institution of Washington. Anolis gundlachii Peters. Eight examples, the largest 176 mm. in total length, from a small settle- ment up the Rio Grande, 2 miles above Arecibo, Porto Rico, on June 12, 1915. B mm. FIG. 4. — Anolis mayeri sp. nov. A, upper surface of head. B, lower surface of head. C, side view of head. D, transverse lepidosis of trunk. E, lepidosis of hind toes, knee, and vent. F, lepidosis of fore-arm and fingers. 10 Papers from the Department of Marine Biology. Color in alcohol uniform dull brownish-olive generally above. Lips pale or whitish, labials obscurely dusted with sooty dots. From gape of mouth back, including ear, whitish band well defined and extending to groin. Immediately below this a gray-brown narrow area all along to groin, fading out in pale color of lower surface, which is largely whitish. Nearly all examples show belly with very pale blue-green tinge. Dewlap yellow-ocher. Some examples are more grayish than others and most show fine specks or dots of dusky on the side above and the costal region below pale lateral streak. The con- trasted and bright markings noticed by Dr. Stejneger in living examples are not evident in these preserved specimens. Dark cross-blotches are, however, seen obscurely, though in smaller examples they are far more noticeable. Scales of the supraocular semicircles usually separated by one row of scales at least, often two rows present, and sometimes partly in contact. Besides the above, 6 examples were obtained from Utuaclo. These often show but a single row of scales separating the supraocular semicircles. Anolis cristatellus Dumeril and Bibron. Figure 5. A very large series, represented by nearly 300 specimens from various local- ities. A large series from Guanica, Porto Rico, shows usually the supraocular semicircles in contact, though the size and disposition of the plates present great irregularity. In color some are very finely vermiculated with darker lines. Often the upper surfaces of the limbs present an obscurely barred 10 K FIG. 5. — Anolis cristatellus Dum6ril and Bibron. Lepidosis of the supraocular semicircles. Figs. A to D and L and M from Guanica, E to H from Arecibo, I to K from Utuado, X and O from Virgin Islands. Amphibians and Reptiles from Porto Rico and the Virgin Islands. 11 appearance, and in some cases they are finely vermiculated. Many examples are more or less olivaceous. In a small series from Guanica one example shows a quite distinct whitish vertebral stripe. Several are pale or light gray. A large series from Utuado, collected both in the town and down the Arecibo River, June 12, are very variable, both in color and squamation of the head. In color these examples appear much more mottled than others. The chin and throat are variable, and may be streaked or uniform, especially in small or young examples. Some show a considerably developed nuchal crest. Often the labials may be marked with dark blotches, giving the lips a cross- barred appearance. Some examples show five or six obscure dusky or blackish saddles on the back. Occasionally a pale vertebral band is seen down the back. Though the squamation of the head is quite variable, usually the supraocular semicircles have two pairs of plates in contact, often one or two small ones may be interpolated in the median line, and occasionally a median series of small scales (in one instance double) entirely separates them. The largest of this series is 175 mm. in total length. Thirteen examples obtained at a small settlement up the Rio Grande, 2 miles from Arecibo, on June 14, 1915. Several show a pale vertebral streak on the back and a longitudinal blackish streak along each side of the back, mostly with narrow pale bordering line. Sides of body may also present obscure irregular dark blotches forming smaller spots or speckled appearance below. In some, the lower surface of the body is yellowish in alcohol. Two from the Arecibo River near the 70-kilometer post, June 14, 1915. One light gray, conspicuously mottled with darker, another with the pale vertebral band bordered each side with a blackish lengthwise band. A large series from the Virgin Islands, many showing the white vertebral band on the trunk, occasionally interrupted. Many have the chin speckled. Frequently a dark spot occurs behind the occipital plate, sometimes divided to form one on each side. Anolis stratulus Cope. Four from Guanica, 2 from Utuado, 3 from a small settlement up the Rio Grande 2 miles from Arecibo, Porto Rico, and 3 from the Virgin Islands. Considerable variation in color is noticeable and a few show distinct dark saddles on the back. All are more or less finely speckled with darker color; in some cases the marbling gives quite a variegated pattern, due to the variable small or large dark spots. The Virgin Islands examples show the back marked with blackish transverse saddles. Dewlap bright yellow in alcoholic speci- mens. Variation in the squamation is often seen. Some specimens show one or two pairs of plates of the supraorbital semicircles in contact. Anolis krugi Peters. Five examples from Guanica, Porto Rico, noAV show the dewlap red. Ten from Utuado, on June 12, show the dewlap crimson in alcohol. Sides and lower surface largely brilliant yellow. In these examples the squamation of the supraocular semicircles is often quite variable. While they may be separated by one row of scales, or sometimes two rows, the rows are frequently incomplete, so that the supraoculars may be in co,ntact with at least one of their pairs. Five from the Arecibo River near the 70-kilometer post, on June 14. These all have but one row of scales between the supraocular semicircles. 12 Papers from the Department of Marine Biology. Anolis pulchellus Dtimeril and Bibron. Seven very interesting examples from the Virgin Islands which agree very well with Dr. Stejneger's description and figures of Porto Rican material. In alcohol these examples are all pale brownish. A dark brownish, well- defined streak alongside of snout below canthus rostralis to eye and contin- ued back from latter well above ear alongside of back, and on trunk more or less broken into spots irregularly to groin. Below ear on each side of lower surface of head a brownish streak to shoulder, though not evident in young. Upper surface of body or back between dark lateral bands with speckling of brownish, mostly obscured, and arranged largely in lengthwise courses. Sim- ilar markings also seen on upper surface of head in some examples. Variation is seen in the scales separating the supraocular semicircles. In one case a pair of plates of the latter are largely in contact, in another instance two rows of scales intervene, while in the other a single complete row occurs. Anolis poncensis Stejneger. Four from Guanica, Porto Rico. This species was heretofore known only from the type locality at Ponce, where it was obtained April 16, 1900. It differs from the other Porto Rican species of Anolis in the completely keeled scales of the back, sides, and belly. In coloration my material agrees with Dr. Stejneger's account, the young or smaller example showing the pale ver- tebral band bordered on each side by a dusky parallel and equally wide band. It also shows a pale obscure orange blotch at and on each side of the occiput, the whole three indistinctly joined. One example is conspicuously greenish above, mottled with dusky. The squamation of the supraocular semicircles is sometimes quite irregular, and these may have two pairs in contact or be separated by a very narrow median row. TEIID.E. Ameiva exul (Cope). Six from Utuado, Porto Rico, some showing the granules extending forward to the second chin-shield. In one an accessory scale is interpolated between the preparietals and in contact with the parietal. One from the Arecibo River near the 70-kilometer post, on June 14. One from a small settlement on the Rio Grande 2 miles from Arecibo, on June 14. It shows two very large plates on the inner tibial surfaces. Among 13 examples from the Virgin Islands, one shows the prefrontals separated by a small accessory scale, and another has the same scale and frontal divided into three equal scales, two of which are in front as a pair; 3 more examples from the Virgin Islands show the occipital plates rounded and subequal; one also shows an additional plate bordering the third and fourth supralabials and interposed between the loreal, first subocular, and preocular. Ameiva wetmorei Stejneger. Frontispiece and Figure 6. Description. — Body elongate, rather slender. Tail long, conic. Head conic, muzzle long. Eyelids granular. Teeth rather large, conic, in one row, and becoming larger as they progress backward. Interorbital level. Ear about size of eye, and tympanum not deep. Hind foot about two-thirds combined length of head and trunk. Front nasal plates meet rather narrowly in contact on top of snout. Nos- tril in suture between two nasals. Supraocular plates three, in contact with prefrontal, frontal, fronto-parietal, and parietal. Fronto-parietals united as a single plate before 3 large parietals, of which median is much narrowest. Suboculars narrow, first two each half length of last. Loreal in contact Amphibians and Reptiles from Porto Rico and the Virgin Islands. 13 with second and third labials, first three suboculars, first supra-preocular, pref rental, fronto-nasal, and hind nasal. First supraocular in contact with first superciliary, loreal, pref rental, frontal, and second supraocular. Second and third supraoculars separated from superciliaries by a narrow series of very small scales. Second supraocular in contact with first supraocular, frontal, fronto-parietal, and third supraocular. Third supraocular in contact with second, fronto-parietal, and parietal. Superciliaries of five plates, last small, about equal and all combined only equal in length to anterior one. A row of narrow occipital plates. Six superior and 5 inferior labials to rictus. Ear-opening not denticulated. First pair of chin-shields in contact, second pair separated only in front by narrow single row of small granules. Three rows of scales on gular fold, front and hind rows of about equal width, median row larger. Scales on upper surface of body and limbs minute. Brachium with one series of very broad plates externally and a narrower inner, series. Antebrachium with three series of small plates. On under surface of tibia two rows of enlarged plates, outer row of much larger plates, and one above enormously enlarged. Eight rows of abdominal plates. Three large anals, one in front and two behind. Tail covered with rings of smooth scales. Color in alcohol largely black above, with 7 longitudinal sharply defined narrow lines of very pale blue-green, distributed as follows: Median, much broader than the others, extends from tip of snout to tail, where it broadens FIG. C. — Ameiva wetmorei Stejneger. A, upper surface of head. B, lower surface of head. C, side view of head. D, transverse lepidosis of trunk in lateral view. E, lepidosis of hind limb and ventral region. F, lepidosis of fore-arm and fingers. 14 Papers from the Department of Marine Biology. out and, like the others, fades in the more bluish color of the tail. A second streak includes the superciliaries and extends back on upper side of head till above shoulder. On each side, from lower edge of eye back over ear, along its upper edge, a narrow pale blue-green line extends back alongside of tail, fading out behind. Another similar line is on each side from lower ear-edge above shoulder back to groin; below this another still narrower line of same color from axillary to groin. Behind hind leg a single pale blue-green line extends back alongside of tail, fading out behind and much wider than lines ending in groin. Upper surface of limbs blackish like back, each forearm with lengthwise median pale blue-green line, broken in one or more places, and continued up on upper arm as several spots. Several pale spots on outer surfaces of limb in black field. Upper surfaces of hind limb conspicuously spotted with pale blue-green. Under surface of body mostly uniform pale blue-green. Feet and hands dark above, pale below. Tail largely pale bluish, especially terminally. Color in life with tail greenish-blue (coppery-blue), ground-color above black and below pinkish gray. Dorsal median longitudinal stripe, and three lateral longitudinal stripes, deep straw-yellow. Shorter stripes begin at both nostrils and extend to groin. Legs spotted yellow on brownish-black ground- color. Total length 103 mm., from snout tip to vent 31 mm. One example from Ensenada, Porto Rico, June 1, 1915. Collected by Mr. M. H. Sanborn. The above is the only example I have seen, though several others have been reported. Its striking coloration will at once distinguish it from the well- known Ameiva exul. Apparently not reported since described.1 AMPHISB^ENID^E. Amphisbsena caeca Cuvier. One from Guanica. Rings on body 217, on tail 17. Long occipitals more than half length of frontals. Length 130 mm. Collected by Mr. M. H. San- born. BOID^E. I'.'pic rates inornatus (Reinhardt). One from Guanica, collected by Mr. Sanborn. Ventrals 264, subcaudals 75. Indistinct blotches are seen all along the body, numerous and most distinct on hind part of body and tail. Ventrals largely uniform deep dusky brown. Length 1,675 mm. CORONELLID^E. Leimadophis stahli Stejneger. Two from Guanica, besides one obtained on a coffee plantation in the mountains 5 miles west of Utuado. They show ventrals 154 to 157, subcaudals 90 to 92, anals 2, scales transversely on back 19. Length 363 to 382 mm. Alsophis portoricensis Reinhardt and Liitken. Two from Guanica, collected by Mr. Sanborn. Ventrals 172 to 180, sub- caudals 116 to 118, anals 2, scales transversely over back 17. Length 534 to 1,057 mm. Alsophis antillensis (Schlegel). Five from the Virgin Islands. They show ventrals 172 to 183, subcaudals 124 to 130, anals 2, scales transversely over back 19. Length 368 to 943 mm. 'Proc. Biol. Soc. Washington, xxvi, 1913, pp. 69-72. Guanica. Amphibians and Reptiles from Porto Rico and the Virgin Islands. 15 Alsophis rufiventris Dumeril and Bibron. One from the Virgin Islands, from which the species has not been pre- viously recorded. Ventrals 212, subcaudals 92, anals 2, scales transversely over back 23. Length 943 mm. EMYDID.E. Pseudemys palustris (Gme'in). A series of 10 very interesting examples from Guanica Lake, Porto Rico. On first examination they appeared to represent two distinct species, and it appears that two very great extremes of color variation occur in Porto Rico. Dr. Stejneger pointed out in 1902 that "there are indications at hand that there may be some constant differences between those inhabiting the different islands, but the material at my disposal is not sufficient to warrant an attempt to separate them." Later Dr. Barbour says : "There are specimens at hand from Jamaica (type locality), Cuba, Haiti, and Porto Rico. They do not, however, vary in such a way as to warrant the designation of varieties confined to each island. Stejneger and I (Bull. M. C. Z., 1910, 52, p. 301) had both expected that such subspecies might be defined. As yet, however, the material available for study is very meager. The species is very shy, and is usually confined to lowland muddy pools. " Thus, as little appears to be known concerning this species, a few notes on the variation are given below. All my examples were secured in the lowland lake at Guanica. Dr. Stejneger's figures 185 and 186 represent what may be considered typical (in a restricted sense) of palusiris. This shows specimens which have the pale lengthwise lines on the sides and lower surface of the head. In this form the carapace is more of a uniform color, without dark blotches or spots, and the plastron may be marked with dark lines along the sutures of the plates, which may vary to very obsolete or even be wanting. About 7 of my examples, representing both sexes of this form, are in the collection. The remaining three specimens exhibit a different pattern of coloration. They also differ in having the front claws a little longer than the longest of the preceding. The carapace in all is decidedly paler over and about the anterior third than the rest of its extent, showing but few markings, and these now obscure, though the sutures of all the plates are conspicuously blackish. The sutures of the plastron and all the lower marginal plates of the carapace are also marked with blackish. The carapace is also largely speckled or blotched with darker. The color-pattern of the head is strikingly different from the other specimens and is shown everywhere with narrow blackish wavy or vermiculated lines, often broken or irregular. The feet are similar. The latter variety was first noticed by Gray as Emys vermiculata,1 and the type locality given as tropical America. I have not seen his subsequent plate,2 though under Trachemys rugosa he gives3 the following description of "Emys vermiculata Gray" : "Shell 7| in. Back brown, closely black-dotted; sternum and under side of margin very closely spotted. Sternal shields dark edged. " The locality is given as the West Indies. Sowerby and Lear figure the Emys rugosa* a fairly good representation of the present variety. The interesting figure by Cocteau5 agrees with the above in the speckled carapace and dark-edged sutures of the plastron. The head is streaked as in the examples here allowed as typical palustris. 'Cat. Tort. Brit. Mus., 1844, p. 25. 4Tort. Terrap. Turt., 1872, plates 42, 43. 2Cat. Shield Rep. Brit. Mus., 1855, plate 13. 6Hist. Nat. Cuba, Sagra, Kept., 1844, plate 2. 3Hand-List Shield Kept. Brit. Mus., 1876, p. 46. FOWLER. PLATE 1 Cfl 4J .e 'C CJ •_ O -o