112 A. D. MICHAEL ON THE LAEVA OF 



We are not without ample information on the subject I am 

 speaking of. The anatomy and life history of the larvaa of compound- 

 Ascidians has been well worked out by a number of eminent 

 biologists, and the literature on the subject is copious. I will only 

 mention two works, viz., Milne-Edwards' " Observations sur les 

 Ascidies-composees des cotes de la Manche " (1844) ; and Eeichert's 

 memoir, " Zur Anatomie des Schwanzes der Ascidien-Larven 

 Botryllus violaceus ") in " Abhandl. d. K. Akad. der Wiss. Berlin," 

 1875. The first-named is a classical work forming the ground-work 

 of much of our present knowledge of the subject, and illustrated by 

 numerous figures, one of which is so like the specimen of the fully 

 developed egg with its larva ready to escape, which is now on the 

 club-microscope, that anyone might suppose that Milne-Edwards 

 had made his drawing from the specimen now before you. The 

 second is an exhaustive treatise on a more special subject, and is 

 illustrated by magnificent plates, which leave nothing to be desired. 



The larval Ascidian is a somewhat tadpole-shaped, free-swimming 

 creature, having a nearly globular body, somewhat truncated in front, 

 and having three conspicuous suckers, mostly cupnliform, at its 

 anterior edge ; behind these there is usually a circle of outgrowths, 

 possibly tentacular. Further back again is the ojDening of the 

 branchial sac, and at one side is a large and conspicuous eye-spot. 

 Behind the body comes a long tail, many times the length of the 

 body, and very singular in formation. It is enveloped externally in 

 a colourless and structureless cuticular test, which surrounds, the 

 more or less circular central portion, and is produced so as to 

 form two broad flat bands set on edge, one on the upper and the 

 other on the under-side of the tail, which run all along it like the 

 fins of a sole, but are much broader in proportion. This test also 

 extends behind the muscular portion of the tail and there expands 

 a little, and is marked with diverging rays, so that it has a certain 

 resemblance in form to the tail of the sole. Within the test is a 

 layer composed of clearly-marked longitudinal rows of large 

 nucleated cells. There are usually eight rows of these cells in 

 Botryllus, but in the species I am now showing to you there seem to 

 me to be only six. The central portion of the tail is a cellular, 

 rod-like body. 



The larva swims with a rapid, somewhat wriggling motion, by 

 means of quick lateral strokes and undulations of the tail, but its 

 free-swimming life is short, usually only a few hours. It then 



