On the Life-History aud Development of the Genus Myzostoma. 559 



comparison may be made without danger of falliug into morphological 

 errors, of the crawling of a Tardigrade. 



The rings of cilia become again broken up. Fig. 25 shows a larva 

 of eight days. It has lost its setae but has not developed any perma- 

 nent limbs. However it appears to possess two somewhat ventrally 

 plaeed processes [p-p] which may aid it in its endeavours to attach 

 itself to a Comafula. It is the only larva with these processes which I 

 have observed, for usually I was unable to keep them living for so 

 long as in this parti cular case. Hence the possibility of its being a pa- 

 thological larva must be borne in mind, and for myself I do not believe 

 it is a normal stage. 



In no case bave I foimd larvae in a free state with permanent legs 

 developed or even with traces of them. On the other band I have found 

 on a Cfomatula a larva (fig. 26) in which only the rudiments of one pair 

 of limbs are present, and these of so small a size that it is inconceivable 

 that they could be of any use to it in finding its way to a Comatula^ or 

 in retaining its place when there. Hence it may be concluded that in 

 the free state no further stages of development are passed through, aud 

 that the rest of the development takes place on the Comafula. This 

 conclusion I think may be accepted in spite of the fact that I have been 

 unable to trace actually the passage from the free state to that on the 

 Crinoid. The difficulties which preveut success in all attempts in this 

 direction are twofold : the extreme smallness of the larvae and the well 

 known difficulty of keeping chaetopod larvae li\àng for many days to- 

 gether. In any case the stages not observed cannot be of any great im- 

 portance, since it is pretty certain that no permanent claws are developed 

 until the larva is on its host. And certainly no other organs are devel- 

 oped tili then. 



And now it becomes a very interesting question as to how the larvae 

 get on a Comaüäa., or even find one at all. It is certain that when in 

 possession of provisionai setae andciliated rings it does not take possession 

 of its host, for if it did, such larvae would be found among the ani- 

 mais which fall off in the alcohol when a Comatida is killed in the way 

 described in the earlier pages of this paper. But no larvae with pro- 

 visional setae are found there, and the youngest found (fig. 26) has 

 traces of one pair of permanent legs. 



The muscular processes figured in the one larva (fig. 25) even if 

 they be such, can hardly be of much service. And indeed I am inclined 

 to think that the larva finds its host in a much simpler manner than by 

 any use of its own limited senses or of its own muscles. Probably the 



