DEVELOPMENT OF A XON-AQUATIC TADPOLE. 257 



expands to some extent (PI. Ill, figs. 1 and 2). A shallow 

 stomodaeal pit arises immediately under the fore-brain. A 

 proctodeum is not formed until later. The fore-gut of the embryo 

 sends downwards a diverticulum {d.f.g.) into the yolk, and from 

 the front of this a prominent anterior cgecum arises which is the 

 beginning of the liver {h, h.c.) 



The vertical diverticulum grows and bends over to the right 

 side of the middle hne. It gradually becomes U-shaped, growmg 

 at the expense of the yolk. A short portion of the middle region 

 of the archenteron tube disappears (fig. 2, a), and a little later 

 the front end of the posterior portion of the archenteron becomes 

 deflected to the left and retains a superficial dorsal position. 

 Thus the gut now consists of a fore and hind-gut which are only 

 connected together by the non-cellular yolk-mass (PI. Ill, fig. 

 8, a.f.g., a.h.g.). 



Growth at the attached or inner ends of the two portions of 

 the g-ut continues, the hind-gut grows forward and sharply dips 

 down ventrally at the front surface of the yolk-mass. The 

 growing end of the U-shaped fore-gut passes up into a dorsal 

 position. Also, the posterior end of the hind-gvit grows backwards 

 so as to project considerably behind the yolk-mass (PI. Ill, fig. 

 4). This is the condition of the gut in a tadpole which has already 

 grown a long tail. 



In the next stage shown in the series both fore and hind- 

 limbs are conspicuous, and the tail has been reduced to a small 

 rudiment. Here the gut is still quite simple, but the whole of it 

 has undergone a profound modification. The fore and hind-gut 

 have become continuous, incorporating the whole of the yolk in 

 the swollen cells. The hind-gut has become stouter and has 

 shortened very greatly (PL- III, fig. 5). From this simple condi- 

 tion the adult configuration is derived, where the j)arts are 

 relatively much longer (Fig. 6). 



Ordinary tadpoles have" relatively a very .much longer gut 

 than in the adult. In the present case, we have a great shortening 

 taking place at the time of metamorphosis, but there is a subse- 

 quent elongation and remodelling of the whole alimentary canal. 

 It has been shown by Bataillon* by direct experiments on living 

 tadpoles that the shortening of the intestine at metamorphosis 

 involves the whole length and not merely some special portion 

 of it; but here there is a profound remodelling of the whole gut 

 after the shortening has occurred. 



There is no trace of external gills, but there is some indica- 

 tion of gill-pouches in the lateral walls of the embryonic pharynx. 

 I could detect no gill-slit grooves in the external epidermis. 



Immediately behind the gill-slit region of the pharynx, the 

 alimentary canal early loses its lumen and becomes solid. A 

 lumen gradually becomes reformed after the fore-limbs have 

 appeared on the outside. 



* Boulenger, G. A. — "The Tailless Batrachians of Europe," Part I, p. 97, 

 Ray Society, London, 1897. 



