igi4 ] S. Kemp : Onychophora. 487 



sequently favourable for the production of young throughout the 

 year. 



Evans has noted that in the case of Eopcripatus it is difficult 

 to imagine how the receptacula seminis obtain fresh suoplies of 

 spermatozoa, seeing that the uteri are completely filled with 

 developing embryos and that young are apparently produced 

 throughout the year He concludes that in Euperipatus fertihza- 

 tion can only take place once during life ; but it is not altogether 

 clear that such an assumption is necessary. When the stock of 

 spermatozoa in the receptacula is either exhausted or, through age, 

 has become powerless, the production of embryos must perforce 

 cease, giving opportunity in due course for the admission of a 

 fresh supply. Naturally, the same difficulty does not arise in 

 the case of Typhlopcripatus, in which fertihzation can be effected 

 annuall}^ at the close of each breeding season. 



The development of the external form in the embryo does 

 not, so far as I have been able to determine from an examination 

 of a limited number of stages ' , offer any very striking peculiarities. 



In the manner of formation of the primitive somites there 

 appears to be a considerable resemblance to Pcripatoidcs novcp- 

 zealandiae'^. The germ-bands develop in a curved line on either 

 side : the}^ are widely separated from one another and between 

 them a ventral protrusion of the yolk-mass is visible (pi. xxxvii, 

 figs. 5,6). I have not found any embryo at all similar to the 

 second stage in the development of Eopenpatus weldoni figured by 

 Evans (Quart. Journ. Microsc. Sci., pi. V, fig. 2, 1902). 



A number of females were found to contain embryos in a 

 comparatively advanced stage, with annulate antennae and all, 

 or nearly all, the limbs differentiated. Two of these are figured 

 in pi. xxxvii, figs. 7-9, illustrating the two dift'erent ways in which 

 the embryo may be folded. 



Affinities 



Following the method adopted by Sedgwick in his concise 

 account of the distribution and classification of the Onycho- 

 phora (Quart. Journ. ]\Iicrosc. Sci.. LII, P- 379, 1908) the 

 principal characters of the genus Typhloperipatus may be thus 

 summarized : — 



1. Number of legs, nineteen or twenty, variable in the same 



species. 



2. Inner jaw with a diastema and saw of denticles. 



3. Legs with four complete spinous pads. 



4. Nephridial openings of the fourth and fifth legs situated 



on the third pad. 



^ In my material the embryos which are well preserved (fixed in hot cor- 

 rosive) represent only comparati\ely late stages. The early stages are not in verj- 

 good condition and, being exceedingly brittle, it was extremely difficult to 

 remove them from the uteri in an entire state. I am thus not able to figure as 

 many stages as I wished. 



} Sheldon Quart. Journ. Microsc. Sci., XXVUI, XXIX, 1888-9. 



