BULLETIN OF THE UNITED STATES FISH COMMISSION. 151 



with that of mauj other vertebrates. Our observations, it is admitted, 

 do not rest upon the evidence presented by sections, but upon the ai)- 

 l)earances of the living transparent objects. The intestine is at first 

 solid anteriorly; its lumen is a mere cajiillary tube on its first ai)pear- 

 ance behind. It does not ai)pear to originate directly from longitudinal 

 folds of the hypoblast which coalese in the middle line below, in some 

 higher forms, but as a nearly solid band of cells just beneath the noto- 

 chord. 



It will be proper to discuss in this place the nature of the jDeculiar 

 vesicle first described by Kupffer, and known as Kupfter's vesicle, but 

 recently renamed the postanal vesicle by Balfour.* It appears in all 

 the cases which we have observed either some time before the closure 

 of the blastoderm, or nearly at the time it closes. It is situated just 

 beneath the tail, between the latter and the yelk. It appears in the 

 Spanish mackerel before the blastoderm closes, and, as far as I can 

 make out, is simply a vacuole filled with fluid, the direct connection of 

 which with the posterior end of the rudimentary intestine, as has been 

 held, has still to be satisfactorily demonstrated. In some forms it per- 

 sists for a considerable time after the closure of the blastoderm, and is 

 so far anterior to the i^oint of closure that it is difficult to see how it 

 can stand in a post-anal relation to the gut, the anal portion of which 

 is developed almost at a point coincident with that where the closure 

 of the blastoderm takes place, and behind the jtosition of the vesicle. 

 Moreover it is usually asymmetrical in position and finally disappears. 

 Its form and position vary also in diflerent eggs, so that I am at a loss 

 to clearly understand its significance. • That it cannot originate at the 

 moment of the closure is^ proved by the fact that in some forms it is 

 present when the blastoderm has covered but three-fourths of the yelk. 

 I have never seen any communication between it and the medullary 

 canal ; however, its further discussion will be resumed when we come 

 to consider forms in which it is more prominently developed. It clearly 

 remains a fact, however, that the anal part of the gut is the first to be 

 developed ; that the oesophagus for a time appears to be a solid band 

 of hypoblast cells below the head, while the point where the mouth will 

 open is not indicated until twenty hours after the young fish has escaped 

 from the egg ; the vent therefore appears about thirty hours before the 

 mouth. 



The notochord eh apijears in embryos eleven hours old as a rod of 

 cells not different in character from those of the other portions of the 

 blastoderm, but shortly afterwards in the region of the trunk of the 

 embryo clearer cells make their aijpearance in the notochord, lenticular 

 in form and arranged transversely to its axis. They may be seen to grow 

 larger and larger until the primitive chorda cells form only thin trans- 

 verse partitions, between which the large, clear cells are developed. 

 Eventually the partitions entirely disappear, when the large, trans- 



Comp. Embryol., ii. j). 61. 



