MUSEUM OF COMPAKATIVE ZOOLOGY. 161 



the distal third which lies within the invaginated portion like a sword 

 within its sheath. The same thing is seen in the spines which are found 

 on each border of the caudal prongs (PL I. Figs. 6, 7, 12).* The tail of 

 the embryo has an entirely different form. Each half of the fork is pro- 

 duced into seven long spines (PL I. Fig. 7). Of these, the three inner 

 correspond to the three internal spines on the tail of the zoea (PL II. 

 Fig. 2). The fourth is the homologue of the prong itself, while the fifth, 

 sixth, and seventh answer to the three minute ones (PL II. Figs. 2, 5, 

 6, 7), which are situated on the outer side of the fork. Curiously enough, 

 the spines of the two stages tend to an inverse proportion, the fourth, or 

 smallest in the embryo, being homologous with the prong of the zoea 

 tail, while the fifth, or lai'gest, is replaced by one of the small external 

 spines (5 ; ) in the subsequent stage. The fourth and seventh are naked ; 

 the rest are fringed with delicate hairs. In a few instances I found the 

 spines of the embryonic skin invaginated in the way already described 

 in the case of the spines of the caudal fin of the zoea. In one example 

 this invagination affected the second, third, and fifth spines (counting 

 from the inside), (PL I. Fig. 6,) in another the third and fifth, in an- 

 other the third only. Without doubt all the longer spines are thus 

 invaginated within the egg-.f 



The two pairs of antennae of the embryo, again, have a much greater 

 development than in the zoea, exceeding in length the swimming-feet, 

 and reaching, when stretched backwards, beyond the base of the abdo- 

 men (PL I. Fig. 3). The first pair (PL I. Fig. 4) consists of a basal seg- 

 ment, within which lies the antennule of the zoea, and which bears two 

 branches, viz. a long one furnished with three longitudinal rows of fine 

 setae, and a very short one. 



The second antennae (PL I. Fig. 5) divide a short distance from the 

 base into two branches, one of which has the form of a simple, blunt, fin- 

 ger-like process (a) ; the other divides again into three branches (1, 2, 3), 

 which are fringed with delicate hairs. In some specimens, at the mo- 

 ment of issuing from the egg, one or more of these branches is in- 

 folded like an inverted glove-finger. The short and blunt process (a) 

 encloses the spinous process (Spence Bate) of the antenna of the zoea, 

 while the triple branch (b), which forms the bulk of the antenna of the 

 embryo, has its homologue in the external branch, or scale (squamiform 



* According to Milne Edwards, the hairs of the new test of adult crahs which are 

 ahout to moult are invaginated in a similar way. (Histoire Naturelle des Crustaces, 

 Vol. I. p. 55. 1834.) 



t Cf. Goodsir, op. tit., PL III. Fig. 17 ; Claus, op. tit., PI. X. Fig. 9. 



