426 MEMOIRS OF THE AMERICAN ACADEMY. 



stomach. In our Tornaria these appendages are both of the nature of lateral plates, 

 and are situated above the intestine. They are formed, as I have shown in a former 

 paper on Tornaria, as diverticula of the stomach; and I have not seen, as Metschnikoff 

 and Keferstein seem to think, the first formation of the water-system from the 

 oesophagus. I merely took it for granted that the odd part of the water-system of Tor- 

 naria was analogous to the odd branch of the Brachiolarian water-system, and that the 

 lateral plates corresponded to the independent branches found in Brachiolaria. The 

 function of these lateral plates remains problematical : they do not become connected, 

 as I formerly supposed, with the odd water-system ; and their subsequent changes in 

 the young Balanoglossus I have not been able to follow. The lateral 

 hollow at first, but eventually their walls become closely pressed together, forming 

 lappets of considerable size, more or less banded and striated, attached along the 

 upper edge and standing off at an angle, like parts of two inverted funnels, from 

 the walls of the stomach, placed one within the other. 



In our Tornaria the water-system is more distant from the dorsal side than in 

 the Mediterranean species, the dorsal pore connecting with it by means of a long 

 tube inserted on one side of the water-system : the dorsal pore also is quite eccentric, 

 and not situated along the median dorsal line, as figured in Metschnikoff's Tornaria. 

 The walls of the water-system are quite stout : the heart, first seen by Fritz Muller, 

 consists of an independent vesicle, situated immediately adjoining the opening of 

 the dorsal canal into the water-system in a sort of depression of the posterior portion 

 of the water-system. In older stages, the heart, which can be distinctly seen to 

 pulsate quite slowly in the earliest stages, is surrounded by an opaque memb 



which does not participate in its pulsations. From the rounded posterior extremity 

 of the water-system project two short sharp spurs riding upon the upper part of the 

 stomach : the water-system is regularly arched laterally, slightly conical at the anterior 



ty where it is connected with the anterior part of Tornaria (the base of the 



eye-specks) by a broad, flat, powerful muscular band which by its contraction 

 can draw in the whole of the anterior part of Tornaria towards the mouth. I 

 shall call anterior, dorsal, ventral, posterior, those parts of the Tornaria which cor- 

 respond ultimately to the anterior, dorsal, ventral, or posterior parts of Balanoglossus. 

 The natural position of Tornaria in the water, while moving, is usually with the 

 eye-specks uppermost. They revolve quite rapidly upon their longitud 



. MM « axis 



and at the same time, inclining this axis, advance by a motion of translation, or 



upon either of the extremities as a fulcrum. Previous to the transformation 



of Tornaria 



quite transparent : the brilliant carmine, violet, or yellow pi 



