Several species of this genus are remarkable for two reasons. First, the great enlargement of the fully-webbed hands and feet, which are then used as parachutes; secondly, the mode of propagation.

Greatly exaggerated notions are, however, entertained about the parachutes, ever since Wallace's description^[110] of the first "flying frog." The creature was brought to him in Borneo by a Chinese workman. "He assured me that he had seen it come down, in a slanting direction, from a high tree, as if it flew.... The body was about four inches long, while the webs of each hind-foot, when fully expanded, covered a surface of four square inches, and the webs of all the feet together about twelve square inches."

The species in question is Rh. pardalis, an inhabitant of Borneo and of the Philippine Islands. Specimens from Wallace's Collection are in the National Collection and the largest specimen shows the following measurements. Total length 6.5 cm. or 2½ inches, not 4 inches.

i.e. for the four limbs 18.8 square cm. = about 3 square inches, and not 78 square cm. or 12 square inches. By some unfortunate oversight Wallace must have mixed up the total expanded area with that of the four hands and feet! In Brehm's Thierleben the 78 square cm. have increased to 81 cm., and the artist has in the somewhat larger species Rh. reinwardti improved upon this, and has produced a truly startling picture by a further exaggeration based upon the figure given by Wallace.

Rh. reinwardti lives in the forests of the mountains of Java and Sumatra. It reaches 3 inches in length, and is grass-green above, yellow below. Younger specimens are further adorned with large blue patches on the webs of the hands and feet and behind the armpits. Besides the flap on the heel and the curious cutaneous fringe on the forearm, suggestive of an incipient flying-membrane, the skin forms a projecting fringe on the inner side of the fifth toe and a transverse flap above the vent.

Of Rh. leucomystax, Annandale, who accompanied the Skeat Expedition to Malacca, gives the following account:–"This frog, which is called by the Malays of Lower Siam either 'Berkata Pisang' (banana-frog) or 'Berkata Rhumah' (house-frog), lays its eggs either on leaves of branches overhanging the water, or on the mud surrounding buffalo-wallows. The ova are enclosed in a round mass of yellow froth, which afterwards becomes steel-grey, about as large as a cricket-ball. Should they be placed judiciously in a position sheltered from the sun, the tadpoles may either hatch, and reach a considerable degree of development, before the mass is washed into the water, or the froth may be melted almost as soon as it is formed and the eggs be carried into a pool by a shower of rain. Very often, however, the whole mass is dried up by the heat of the sun before the rain comes. During the breeding season, which seems to occur as often as the land is flooded under the trees, for I have never seen the eggs of this frog on the bank of a river, the males croak loudly, producing a sound which can hardly be distinguished from the chattering of the large black and yellow squirrel, Sciurus bicolor."

These arboreal frogs have a peculiar mode of nursing the young and taking care of the eggs. Rh. maculatus of Ceylon, Malacca, etc., and Rh. schlegeli of Japan, lay their eggs in a foamy mass, the size of a fist, on the margins of ponds, and the whole process has recently been described by Ikeda.^[111] He observed the Japanese Rh. schlegeli depositing the eggs in soft, muddy ground covered with grass, and in wet, muddy banks of paddy-fields, ponds, and similar localities near Tokyo. Sometimes they are deposited between the leaves of trees, near the ground. The breeding season extends from the middle of April to the middle of May. Towards the evening the female, bearing the much smaller male on her back, retires underground for the deposition of the eggs. The spots chosen are 10-15 cm. above the surface of the water; the female digs a spherical hole 6-9 cm. wide. Sitting thus concealed underground, the frogs assume a dark colour and the spawning takes place during the night, whereupon the parents leave the nest. The eggs are enveloped in a white mass of jelly full of air-bubbles, the whole frothy lump looking like the well-beaten white of a hen's egg, with the yellowish eggs scattered through it, and measuring some 6 cm. in diameter. The air-bubbles are 2-3 mm. large. The froth is originally very elastic and sticky, but it gradually sinks down, becomes liquid and ultimately runs out of the hole. It is produced in the following peculiar manner. During and after the deposition of the eggs the female puts her feet upon the sticky jelly, part of which adheres and is then pulled out as a thin, transparent membrane stretching between both feet. The latter are then thrust backwards, the membrane is folded downwards and becomes a vesicle of 5 to 10 mm. in width. By repeated working of the limbs the successively formed bubbles are trodden and kneaded into froth, which ultimately surrounds and at the same time separates the eggs.

The female of Rh. reticulatus of Ceylon attaches the eggs, about twenty in number, to the under surface of her belly, on the skin of which they leave little cellular impressions. What becomes of the tadpoles is not known.

Rh. leucomystax is found in the Malay Archipelago, Farther India, and the Philippine Islands.

S. S. Flower^[112] found the tadpoles about Singapore, from January to April, in small ponds and in rain-water butts. The spiracle lies on the left side, directed backwards and upwards, nearer the anus than the end of the snout. The anus opens on the right side. Exceptionally large tadpoles measured 46 mm. in total length, the recently transformed young only 14-18 mm.

"A cheerful little frog of most graceful build. It comes out from its hiding-places shortly before sunset, and remains abroad all night. The males are easily found as they sit on shrubs or trees, or on the edges of the rain-water butts under the verandahs of the houses, and from time to time utter a single, rather musical, short croak. In March and April they can be found both by day and night in embrace, in the ponds. This species changes both its colour and markings very rapidly and frequently, but dark bands across the legs can always be more or less distinguished; the lower parts are some shade or other of buff, but the principal variations of the upper part are as follows: pale bronze, either uniform or with four longitudinal dark-brown or black lines; uniform, almost orange, bright bronze; chocolate, with darker mottling; pale brownish green or olive, with irregular dark spots; yellowish green, mottled with darker or brown." The females are considerably larger than the males; the largest male caught was 48 mm. from snout to vent, and the largest female 68 mm.

Rana.–The following combination of characters should be a sufficient diagnosis: pupil horizontal; tongue deeply notched and free behind; vomers with teeth; fingers free, toes webbed, fourth and fifth metatarsals diverging and webbed together.

In conformity with the great number of species and the wide distribution of this genus some of the organs vary considerably, indeed so much so that many of these modifications have been deemed sufficient to be of generic importance. Fortunately the species are so numerous that these characters mostly form an uninterrupted series from one extreme to the other.

The terminal phalanges are mostly simple and pointed; sometimes transversely dilated or T-shaped, according to the presence of more or less developed discs. Such discs are, for instance, present in the Malay species R. erythraea and R. chalconota and in the Indian R. corrugata. The tympanum occurs in every stage from a conspicuous, free disc to being quite hidden by the skin. The vomerine teeth either form a pair of tiny, mostly transverse rows, between the choanae, or they are arranged in two oblique series which extend beyond the hinder edges of the choanae.

The vocal sacs vary greatly. Many species, e.g. R. agilis, have none at all. Most species have a pair of internal sacs, and in comparatively few, about a dozen, these sacs have become external, a feature which indicates no relationship of the species thus distinguished, for instance the European R. esculenta, the Japanese R. rugosa, the Indian R. hexadactyla, R. cyanophlyctis and R. Chloronota, the Bornean R. glandulosa, the African R. oxyrhynchus and R. mascareniensis, the Mexican R. montezumae. In R. esculenta, and perhaps in a few others, even the female has some traces of these otherwise male organs, indicated by slit-like folds of the outer skin below the angles of the lower jaw.

Nuptial excrescences on the inner metacarpal tubercle and on the inner fingers of the male are common; they reach their greatest development in the Himalayan R. liebigi, the male of which is "remarkable for the extreme thickness of its arms, the inner sides of which are studded with small conical black spines, each supported on a rounded base produced by a swelling of the skin. A large patch of similar spines exists on each side of the breast."^[113]

Specific glandular complexes in the skin are mostly restricted to a pair of lateral or dorso-lateral folds; they are often absent, but a few species, e.g. R. glandulosa of Borneo, R. temporalis of Ceylon, R. elegans and R. albolabris of West Africa, have a pair of large flat glands at the base or inner side of the arms.

All the species of Rana, except those in the Solomon Islands, spawn in the water, where the development of the tadpoles takes its course. Those of some Indian species, notably R. alticola and R. afghana of the Himalayas, and R. curtipes of Malabar, are very peculiar, being provided on either side of the shoulders with a large oval parotoid-like gland, well defined and crowded with pores; R. alticola possesses in addition an unpaired, sharply marked glandular complex on the top of the root of the tail, or rather upon the future coccyx. These complexes gradually disappear with age.

The genus Rana, with about 140 species and subspecies, is distributed over the whole of Arctogaea so far as this is available for Amphibian life, while there are only a few stragglers in Notogaea, namely, a few species in Ecuador and in the Peruvian or Upper Amazon district. None exist in the rest of the Neotropical region, including the Antilles, and practically none in Australia; but R. arfaki and R. papua inhabit New Guinea and the northern corner of Australia, R. kreffti the Solomon Islands. A few species are restricted to Madagascar, and a few others live there and on the continent of Africa.

So far as number of species is concerned, the home of the genus Rana is the Palaeotropical region; about one dozen (some of them with a very wide range) live in the Palaearctic sub-region, scarcely more in the Nearctic sub-region, and a few in Central America.

R. temporaria (the common European Brown Frog or Grass-frog).–The tympanum is distinct, two-thirds the diameter of the eye in size. The first finger is slightly longer than the second, which is shorter and weaker than the others, whilst the fourth is the longest. All the fingers are quite free. When the hind-limbs are laid forwards along the body, the ankle-joint reaches to a point between the eye and the tip of the snout. The five toes, which are about half webbed, increase in length from the first to the fourth, while the fifth is about equal to the third. The sole of the foot has a small, blunt, inner metatarsal tubercle; the outer one is scarcely visible. The skin is smooth, always moist, owing to the minute mucous glands; but a series of larger glands forms a pair of folds along the upper sides of the back; beginning behind the eyes they converge slightly beyond the shoulders, diverge a little in the sacral region, and converge again towards the vent. Another, much feebler, Λ-shaped ridge lies between the shoulders.

The male has two internal vocal sacs, which, when in use, bulge out the skin of the throat beneath the angles of the mouth like a pair of globes. It is further distinguished from the female by the stronger muscles of the arms and by a pair of swollen pads on the inner side of the first finger. During the pairing season these pads are enlarged into cushions covered with black horny rugosities.

The iris is golden, with dark specks. The coloration is, generally speaking, brown above, with black-brown irregular spots, especially on the sides of the body, and with cross-bands on the legs. The under parts of the male are white or pale yellow, with a bluish tinge on the throat, while the female is more yellow instead of white, inclining to orange. In both sexes the under parts are mostly spotted with darker colours. A large dark-brown patch, extending from behind the eye over the tympanum towards the shoulder, is always present and has given this frog its specific name. Otherwise the coloration varies considerably; more or less according to the locality and nature of the surroundings, and to individual variation and temporary change of colour.

Some specimens are almost spotless above and of a rich brown, or almost yellow colour, the spots being restricted to the sides below the lateral folds. Others have very few spots, but these are then arranged in two interrupted streaks on the back. The under parts, especially the flanks, may be lemon yellow instead of whitish, and the darker markings may be almost absent. Boulenger has figured a beautiful specimen, almost orange red, with red spots and vermiculations on the yellow under surface. I have found similar red specimens of unusually striking appearance between Berlin and Spandau in a forest-glade, through which run little streams with banks of red ferruginous soil. Specimens which live in woods with rich black soil are often very dark, all the brown and reddish tints being absent. The variations are, however, really endless, and it is difficult to find two individuals exactly alike, even amongst a great number collected in the same locality. Moreover, they change colour. Warmth makes them paler, cold causes the chromatophores to expand and the whole frog appears darker. During the breeding season the males assume a delicate bluish hue, especially on the throat, but this film quickly fades away when they are taken out of the water. It is caused by the swelling of the cutaneous lymph-spaces which extend their ramifications into the epidermal layer, and it is not a question of pigmentation or of chromatophores, but a case of interference-colours, blue being frequently the result of the light passing through a cloudy, colourless, but not quite transparent and thin stratum, in this case the turgid epidermis.

The habits of the Grass-frog are essentially terrestrial. It spends most of its time on land, preferably in damp places, but local fashion permits of a great deal of freedom, as these frogs are sometimes found not only in very wet, naturally irrigated places, but also in the water itself. However, the Grass-frog when pursued rarely takes to the water for safety. It trusts to flight, first by a few long and fast jumps, and then to concealment by squatting down between grass, under leaves; it rarely creeps into a hole, even if there be one near. The jumps soon become shorter and shorter after a few dozen repetitions. It swims well, but cannot climb. The food, which consists chiefly of insects, snails, and worms, must be moving to excite interest; then the frog, whose favourite position is half squatting, half supported by the arms, erects itself, and, facing the insect, turns round upon its haunches, adjusts its position anew by a shifting of the legs, and betrays its mental agitation by a few rapid movements of the throat. All this time the prey is watched intently until it moves; then there follows a jump, a flap of the tongue and the insect is seen no more. As a rule these frogs do not crawl, they jump or hop, even whilst stalking, and this takes place at any time of the day; in fact they are very diurnal, although they become more active towards the evening. When caught they are at first very wild and, like all true frogs, very impetuous, committing acts of astonishing stupidity without any apparent sense or appreciation of distance or height. The captive will not only jump off the table, whilst a toad stops at the edge and looks carefully down, but without hesitation he jumps out of the window, regardless of the height above the ground. This is due to sheer fright; he loses his head. When at large in his native surroundings, nothing will induce him, although hotly pursued, to commit suicide by jumping down a precipice. But all this wildness calms down wonderfully soon. The captive no longer dashes his head against the glass, he does not struggle or twist when taken up; on the contrary, he makes himself at home, watches your coming with intense expectation, and without hesitation accepts the proffered mealworm, maggot, butterfly or earthworm; in short, he shows what a jolly and intelligent fellow he really is.

The Grass-frog has many more obvious enemies than perhaps any other Amphibian, and it is not even slightly protected by any appreciable poisonous secretion. Nevertheless it is extremely common. A whole host of birds eat it–for instance, buzzards, harriers, and above all storks. Foxes, polecats, and stoats are not averse to it, and the Grass-snake derives its main sustenance from it. In fact the enemies of the little frog are legion, one of the worst being Man. In France, Italy, and other parts of the Continent, the skinned fleshy hind-limbs are turned into a by no means disagreeable ragoût, or into dainty morsels when fried in butter and encrusted with bread-crumbs. This frog, together with its cousin the Water-frog, also suffers from the distinction of being one of the chief martyrs to science. Perhaps the story is true that Galvani was led to his investigations into animal magnetism and electricity by observing that the legs of a number of skinned frogs, strung up by his wife upon the bronze railings of the balcony, jumped whenever the scissors, which cut off the feet, touched the other metal. Frogs have suffered ever since. Easily procured and of a convenient size, they are used in every biological laboratory, and the young student is supposed to be initiated into the mysteries of Vertebrate structure by the careful dissection and study of this, the worst of all the so-called types. Next to Man there is no animal which has been studied so minutely, and has had so many primers and text-books written on it, as this frog. In spite of all this it is very little understood, thanks to its rather aberrant and far from generalised structure.

However, the frog, by reason of its fertility, holds its own. Early in the year, sometimes while there is still ice and snow, the frogs leave their hibernating places (mostly holes in the ground, under moss, or in the mud), and they begin to pair in standing or slowly flowing, mostly shallow, waters.

They are not always very careful in the selection of the spawning locality, many of them lay their eggs in a ditch, or even in the shallowest puddle, which is sure to dry up, and thus to cause the destruction of the whole brood. This carelessness is all the more surprising when there are large pools or lakes in the immediate vicinity, perhaps only one hundred yards to the other side of the road. The Natterjack is, by the way, equally careless, while other toads and the tree-frogs are very circumspect.

Both sexes can croak, and this sound is frequently produced under water; but there are no regular concerts, although many collect in the same pond or spring, which is perhaps the only suitable place for miles around. The male puts its arms around the chest of the female, behind her arms, and the embrace is so firm that nothing will induce him to loosen his hold. The process becomes an involuntary reflex-action, a cramp which may last for days, or even for weeks, if sudden cold weather sets in, until the female is ready to expel the eggs, an act which is quick and soon over. The usual time of spawning in Middle Europe is the month of March, earlier in warm, later in cold seasons; in southern countries, February or even January, but in Norway not until May. Although the males of this species are not more numerous than the females, and therefore should be able to mate without much trouble, their ardour is so great that they occasionally get hold not only of the wrong kind of frogs, but of toads or even fishes, and, if taken off by force, they fasten on to anything else, a log or on to your own fingers. The eggs measure 2-3 mm. in diameter, are black with a whitish spot on the lower pole, number from 1000 to 2000, and sink at first to the bottom. Their gelatinous cover soon swells to a large globe more than 10 cm. in width, and the whole mass, as large as a man's head, floats on the surface, often stained with mud and other impurities. During the cold weather which often prevails in the spring, the dark brown larvae are slow in their development; and provided with rather large branched external gills and a well-developed tail, they wriggle about in the dissolving slime for three or four weeks. Fischer Sigwart^[114] has timed and measured them as follows.–The eggs were laid on the 10th of March. On the 15th the larvae were 4 mm. long and began to leave the eggs. On the 19th they measured, body 4, tail 9, total 13 mm.; on the 5th of April 10, 16, and 26 mm. respectively. On the 13th of May they were 40 mm. long and the hind-limbs appeared; the fore-legs burst through on the 25th, when the tadpoles had reached their greatest length, namely 45 mm., the body measuring 15 mm. On the 31st of May they left the water, still provided with a rather long tail of 20 mm., the total length being reduced to 35 mm. The larvae of this set developed unusually fast, perhaps owing to artificial conditions. The whole development is, however, mostly finished in three months, so that the little stump-tailed baby-frogs swarm about well before midsummer, and have time enough to grow to the size of 20 mm. or ¾ inch before they begin to hibernate in October.

In higher localities and in northern countries the tadpoles are sometimes obliged to winter in the unfinished condition.

In spite of the unusually hot summer of 1899 I found plenty of tadpoles on the 10th of September in the tarns of the hills of North Wales, 1500 feet above the level of the sea; while thousands of little frogs, with and without stumpy tails, were hopping about in the surrounding bogs. The water of these tarns is always very cool. Cold and rainy weather set in by the middle of the month, and on the 26th the tadpoles, all rather small, measuring only 35 mm., with the four limbs developed, but still with a broad fin on the tail, had all settled down under stones at the bottom of the now very cold water, prepared for hibernation. A few were taken home and kept in a glass vessel with water, cool, but less so than that of their native tarns. Within two days they lost the fins on their tails; before the end of a week they left the water, and crawled on to the moss, and the tails were reduced to little stumps. By the 10th of October the metamorphosis was complete, the little frogs measured only 13 mm. in length and showed no desire to hibernate in the genial atmosphere of the greenhouse.

This species has a very wide distribution. It ranges from the west of Ireland to the islands of Saghalin and Yezzo, being found everywhere in the enormous stretch of intervening countries, practically the whole of Central and Northern Europe and the middle belt of Asia. Its most northern extent is the whole of Sweden and Norway. I have found it to the east of the Dovrefjeld, at an elevation of 4000 feet, well-nigh the snow-line. In conformity herewith it ascends the Italian Alps up to 10,000 feet. The southern limit in Europe is the Cantabrian range and the hilly province of Galicia. In the rest of the peninsula, in Italy and Lombardy, Greece and Turkey, and on the Mediterranean islands it is absent.

R. arvalis is often confounded with R. temporaria, as it differs from the latter only by the following characters. The snout is rather more pointed and is narrower; the inner metatarsal tubercle is large, compressed, and hard; the dorso-lateral glandular folds are more prominent and the belly is white and immaculate; lastly, it scarcely reaches 3 inches in length, a size which is not rarely surpassed by the other species. There are also some differences in habits. R. arvalis prefers moist, boggy, open localities, and does not ascend beyond 2000 feet in Central Europe. It pairs as a rule later in the spring and the eggs are smaller, only 1½-2 mm. in diameter; they do not swell up so much, and the whole mass does not float but remains at the bottom of the shallow water. The coloration much resembles that of R. temporaria, and is likewise subject to much variation, except that the pale vertebral stripe is perhaps more common. This species is distributed over the whole of Central Europe, Russia, and Western Siberia, south of the 60th degree of latitude, living side by side with R. temporaria. Between the rivers Elbe and Rhine it becomes decidedly rare, and the latter river is practically its western boundary, while the Bavarian Alps and the Danube form its southern limits.

R. agilis is still more frequently confounded with both the two former species. It differs from either by the absence of the two internal vocal sacs of the male, and by the decidedly longer hind-limbs, the tibio-tarsal joint reaching often a little beyond the tip of the snout. The inner metatarsal tubercle is as prominent as in R. arvalis. Total length up to 3 inches. The prevailing colour of the upper parts is rather yellow or pink-brown with few and small blackish spots; a Λ-shaped dark mark on the neck is often present, and the large dark patch on the temporal regions is always conspicuous. The under parts are white, inclining to lemon yellow on the flanks and thighs. The iris is golden yellow in its upper half, dark brown in the lower half.

This species has a much smaller range than the first two:–from France through Middle and Southern Germany, Switzerland, and Lombardy to Hungary and Greece. The specific name refers to the quick and long leaps of this pretty, or rather delicately coloured frog, which prefers woods and wooded glens to large open places. Their voice differs much from the croak of the common Brown Frog, and agrees with that of R. arvalis, which is transcribed by Boulenger, who has kept them alive, as a rapidly uttered "co-co-co," or "cor-cor-cor." According to the same authority, the pairing takes place as in R. temporaria, but is of much shorter duration, the females usually resorting to the water only at night and when quite ready to spawn. Specimens in embrace are therefore seldom found in the daytime. The eggs resemble those of R. temporaria in size, but they do not swell up so much and they do not float.

These three species of European brown frogs, difficult enough to distinguish, have of late been increased by three more, thanks to the sagacity of Boulenger. These latter inhabit South Europe, and the males all lack the internal vocal sacs.

R. iberica has a very small range, namely the north-western portion of the Iberian peninsula, from the Tagus northwards into Galicia, but south of the main extension of the Cantabrian chain. The rest of the Peninsula south of these mountains has no brown frogs, the only species of Rana being R. esculenta. R. iberica is rather local, being restricted to those hilly and mountainous districts which are well watered. A favourite haunt is the numerous streams in the wooded parts of the Serra Gerez, the red, disintegrated granite of which suits this little, extremely active, and reddish frog to perfection. The prevailing ground-colour varies according to the district, from pale to dark reddish or orange brown, with red specks and larger, dark brown spots, which in some specimens begin with the Λ-shaped mark between the shoulders. Dark spots on the flanks are very variable; the hind-limbs show the usual darker cross-bars, and the temporal region has the conspicuous dark patch. The ground-colour of the under parts is whitish, suffused with a pink tinge, and the throat is much speckled with brown; the toes are pink. The size of this pretty frog amounts to 2 inches. The breeding time is the month of March. When caught and squeezed they emit a slight "co-co-co."

R. graeca inhabits Italy and the Balkan peninsula from Rosina to Morea, together with R. agilis, from which it is very difficult to distinguish except that it is a little smaller, remaining below 2½ inches, and is generally more uniformly pale grey brown to yellowish and pinkish brown above, with scarcely any, or only a few, small dark specks on the back and limbs. The temporal patch is likewise paler than in the other species. The flanks are spotless, their colour gradually passing into the light buff of the under parts, which are more or less marbled with grey. The iris is golden, speckled with dark brown.

R. latastei of Lombardy and Northern Italy down to Florence is the last of these closely allied frogs. Its affinities lie with R. iberica and R. agilis. The latter and R. latastei, although living side by side in the same locality, for instance near Turin, are said not to interbreed. The voice is a rapidly uttered "keck-keck-keck;" the length remains below 2½ inches. The ground colour is greyish or reddish brown with a dark brown Λ-shaped mark between the shoulders, and a few red, orange, or blackish spots on the back. The flanks are without definite dark spots. The under parts are whitish, with a strong pink tinge, especially along the middle of the throat and on the chest, the paler portions being mottled with pale grey brown.

Perhaps the least unsatisfactory way of distinguishing between R. agilis, R. graeca, and R. latastei (R. iberica need not be confounded with them on account of its distribution) is the size of the tympanum, and its distance from the eye. The tympanum is smallest in R. graeca, its diameter being about half that of the eye and from ¾ to the whole of its width distant from the eye. In R. latastei the tympanum is a little larger, and about ½ to ⅔ its own width distant from the eye. R. agilis has the largest tympanum, measuring about ¾ of the diameter of the eye, and the distance between the two organs amounts to only ⅓ of the size of the tympanum.

Brown land-frogs of the R. temporaria group are found in most countries of nearly the whole Periarctic and Oriental regions, and by the time their races and varieties have been studied as minutely as those of Europe are now being scrutinised, the number of species will indeed be great.

R. silvatica is the chief representative in North America. It closely resembles R. agilis, but is smaller, only 2 inches in length, and possesses a pair of internal vocal sacs. Its specific name refers to its predilection for forests of oak, among the dried leaves of which it conceals itself so successfully that it is discovered with difficulty. R. japonica of Eastern Asia is almost indistinguishable from this American species and from the European R. agilis.

R. opisthodon of the Solomon Islands has the vomerine teeth in two oblique series entirely behind the level of the choanae. The general shape of this large frog is stout, the type specimen of the male measuring 78, that of the female 125 mm. = 5 inches. The upper surface of the female is covered with small, flat warts, that of the male is much smoother. The upper parts are dark brown, while the under surface is brownish white. The male has two internal vocal sacs.

This species is interesting as affording another instance of shortened development, the whole metamorphosis being gone through within the egg. Mr. Guppy, its discoverer, has supplied the following notes: "During a descent from one of the peaks of Faro Island I stopped at a stream some 400 feet above the sea, where my native boys collected from the moist crevices of the rocks close to the water a number of transparent gelatinous balls, rather smaller than a marble. Each of these balls contained a young frog, about 4 inches in length, apparently fully developed, with very long hind-legs and short fore-legs, no tail, and bearing on the sides of the body small tufts of what seemed to be branchiae. On my rupturing the ball or egg in which the little animal was doubled up the tiny frog took a marvellous leap into its existence, and disappeared before I could catch it. On reaching the ship an hour after, I found that some of the eggs which I had put in a tin had been ruptured on the way by the jolting, and the liberated frogs were leaping about with great activity. On placing some of them in an open-mouthed bottle, 8 inches long, I had to put the cover on, as they kept leaping out."

Boulenger^[115] has figured and further described the eggs and young. The egg measures 6-10 mm. in diameter, and is a transparent capsule in which the young frog is coiled up in the same way as figured by Peters in Hylodes martinicensis; but none of the specimens, which are in an advanced stage of development, show anything of a tail. There are no gills, but on each side of the abdomen are several regular transverse folds, the function of which is probably that of breathing organs, like the tail of Hylodes. The tip of the snout is furnished with a small conical protuberance projecting slightly through the delicate envelope of the egg, and evidently used to perforate that envelope.

R. guppyi, likewise an inhabitant of the Solomon Islands, is a giant among frogs. It was discovered by Mr. Guppy on the Shortland Islands. The type-specimen measures 165 mm. = 6½ inches in length! The skin of the upper parts is covered with minute warts, and forms a strong fold above the distinct, but small, tympanum. General colour dark olive brown above, dirty white below.

R. tigrina is a common species of Eastern Asia, including the Malay Islands. On account of the strength of its voice, and its size, which is said to reach 7 inches, it is called the "Indian Bullfrog." Mainly aquatic, it has a strong cutaneous fringe along the outer side of the fifth toe. The skin of the back is thrown into longitudinal folds, and a strong fold marks the upper border of the tympanum. The general colour above is olive brown, with dark spots, often with a light vertebral line; the under parts are white. The male has a pair of large external vocal sacs.

R. gracilis has the same distribution, but it remains much smaller, and the toes are only half, instead of fully, webbed.

R. catesbiana is now the settled name of the "Bullfrog" of North America, the much more appropriate name of mugiens having been sacrificed to the fetish of priority. The tympanum is extraordinarily large, at least equal to the size of the eye, largest in the male. The first finger does not extend beyond the second; the toes are connected by a broad web down to the ends, and there is a small inner, but no outer, metatarsal tubercle. The upper parts are olive brown, clouded with dark brown or blackish spots; the under parts are yellowish white, often marbled with brown, especially on the throat. The iris is reddish, with an outer yellow ring. The male possesses two internal vocal sacs. Total length of adult specimens about 5 inches, but there are giants on record 7 inches in length, while the stretched hind-limbs measure another 9 or 10 inches. Its home extends over the whole of the United States, East of the Rocky Mountains, southwards into Mexico, northwards into Canada.

According to Holbrook the Bullfrogs are solitary in their habits, only collecting together in the breeding season, when hundreds may be seen in the same small pond; and then the croak uttered by the males is so loud as to resemble the distant roaring of a bull, and can be heard on still evenings at a distance of half a mile. The voice is a hoarse bass "brwoom," playfully translated into "more rum." "They cannot be said to abound, but are found commonly enough sitting half immersed in water, or on the banks of ponds, waiting for their prey. If alarmed they hop suddenly into the water, but do not conceal themselves at once, frequently skimming along the surface for several yards before they dive below." They are the most aquatic of all the North American frogs, and Holbrook has known specimens to live in wells for years, where they could not rest a moment on solid ground above the water.

The Bullfrog is voracious, and takes almost anything that lives or gets into his own pond–Mollusca, Crustacea, fishes and, above all, frogs. There is no doubt that they drag down and swallow a good many ducklings and the young of other water-fowl, but certainly not the half-grown birds which have a way of disappearing from the farms wherever there are negroes and other farm-hands about. In turn the bullfrog has sufficient enemies to keep its numbers down, in fishes, birds, otters, and snakes, and, in the South, alligators. Although easily kept and growing comparatively tame, they are dull, having to be kept in solitary confinement on account of their greediness, which knows no limits. Two of our specimens each swallowed a full-grown Salamandra maculosa, and died within the same night, probably not understanding the meaning of the conspicuous black and yellow warning colours of the European.

E. clamata s. fontinalis, likewise an inhabitant of Eastern North America, may be called a smaller edition of the Bullfrog, its usual full-grown size being about 3½ inches. The tympanum is conspicuously large, but the toes are webbed to a lesser extent, and the skin forms a glandular fold which extends from the shoulder in a curve to the flank. This species is partial to the neighbourhood of running streams; it is said to be exceedingly timid, and to utter a short cry when disturbed and making its enormous leaps.

Another North American relation is R. halecina s. palustris, frequenting the neighbourhood of ponds and rivers, very lively and capable of jumping 8 to 10 feet. The tympanum is smaller than the eye, but there is the same glandular lateral fold as in R. clamata. The vocal sacs are internal and decidedly small.

R. esculenta.–The common Water-frog of nearly the whole Palaearctic region is closely allied to the American Water-frogs described above, and, like most of them, has the vomerine teeth in two small oblique rows between the choanae and extending a little beyond their posterior border. But the males have a pair of external vocal sacs. The tympanum is distinct, about two-thirds the size of the eye. The first finger is slightly longer than the second. The toes are entirely webbed. Besides the usual subarticular phalangeal tubercles, the sole of the foot is provided with two metatarsal tubercles, the outer of which is very small, while the inner is much larger, although varying in size from a soft oval to a long, curved, shovel-shaped structure. The skin is smooth, except for a pair of prominent glandular folds which extend from behind the eye along the dorso-lateral line. The coloration varies considerably. The upper parts are mostly greenish brown, with black brown spots on the back, and larger patches on the limbs. Most specimens have three lighter stripes along the back, the middle one mostly green, the two lateral bronzy brown and coinciding with the glandular folds. The tympanum is brown, and there is occasionally a dark temporal patch. The posterior aspect of the thighs is invariably spotted with black and white or yellow, in opposition to the R. temporaria group, where these parts are never spotted.

The total length of this species varies much. Specimens 2½ inches in length are certainly mature, those of 4 inches are unusually large, and Boulenger has received a giant from Damascus, which measured 125 mm., or nearly 5 inches. The females are larger than the males.

The variations in colour are not only local but also individual, moreover the colours are changeable. The ground-tint ranges from dull brown through olive to bright green, the dark spots being more or less pronounced and numerous; the light vertebral line is olive-yellowish, bright green, or altogether absent.

Those which inhabit waters with plentiful vegetation, like water-lilies and other luxuriant plants, are generally prettier and more vividly coloured than those which live in swamps and ponds with dark mud, or where the prevailing vegetation has a sombre aspect. Cold and dull, warm and sunny days also influence the water-frogs, and those which have been kept in a dark tank look very different from the bright assembly which had been put in some weeks before.

Various attempts have been made at subdividing R. esculenta of Linnaeus into sub-species, and Boulenger has now, after the attentive study of an enormous material, arranged them in four principal and recognisable races. The chief differences are the relative length of the femur to the tibia and the size of the metatarsal tubercles.

1. Var. ridibunda, Pallas.–The right and left heels overlap each other when the thighs are stretched out at right angles to the vertebral column, and the tibia is closely folded up against the thighs. When stretched forwards, the heel reaches the eye or even the tip of the snout. The inner metatarsal tubercle is feebly developed, very small and blunt; the outer tubercle is absent.

That part of the thighs which is concealed by the legs when the animal is at rest is whitish or pale greenish, marbled with dark olive, or bronze, or of the latter colour with or without small light spots. No trace of yellow is ever to be detected on that region, nor at the axillae or on the groin. The vocal sacs are strongly pigmented with black, when inflated they are pale grey. The iris is a mixture of black and gold.