EARLY STAGES OF TEREBRATULINA. 33 



made out. As this could not be verified in other specimens of the same age, the figure is 

 not given. 



A still more advanced stage is shown in fig. 21. The shell is now becoming propor- 

 tionately broader, and the cirri increase in number, though still forming a simple circle 

 around the mouth. The crura have also begun to form. 



In fig. 22 a con.siderably more advanced state is shown. The valves have been forcibly 

 separated, and the smaller one is thrown back. The cirri are more numerous, numbering 

 thirty-one ; two of them are seen encroaching upon the circular lophophore, and at this 

 stage the lophophore has begun to assume its hippocrepian character. 



The crura (cr) are plainly seen supporting the crown of cirri, and the liver already 

 shows the first indications of its differentiations into the peculiar ca^cal ramifications which 

 become so numerous in the adult ; and at this stage is seen the division of each lateral 

 portion of the liver into a dorsal and ventral lobe ; a lateral lobe is shown in fig. 23. 

 It will be seen that the liver is divided into a series of caeca, though these are united. 



The divaricators (d) are completely foiTaed, and between these two muscles is seen the 

 lengthened intestine {i), the blind extremity of which is held finnly to the shell by a 

 membrane, called by Hancock the ventral mesenter}'. 



The appearance of the shell at this stage is represented in fig. 26. Radiating ribs, to 

 the number of fifteen, ornament the shell. It will be seen by this figure, that the setae 

 correspond in growth and position to the radiating furrows, and this observation is also 

 made by Hancock in his examination of the adult. A well defined concentric line seems 

 to indicate the earlier lingula-shaped shell, and though no ribs appear in the earlier stages, 

 they are defined upon this area afterward, and it would appear from this that the ribs are 

 formed on the inner surface, and that the seta3 direct and induce the furrow. The lobes 

 of the liver (?), and the peduncle (j?) and peduncular capsule, are shown in this figure. 



In fig. 24 a stage stiU more advanced is given ; the valves are thrown back, disclosing 

 the stomach and inte.stine suspended from the calcareous loop. The crura (cr) are well 

 defined, and the ventral mesentery {v7n) is more distinctly seen in this figure. Two 

 lateral bands are seen holding the stomach in position ; and these appear to be attached 

 to the crura, though they probably pass by them, and become attached to the ha^nal 

 valve. These are the lateral gastro-parietal bands fir.st described by Huxley. 



The walls of the blind intestine are yet light brownish in color, as in the Phylactola?- 

 mata, and, as before remarked, are hepatic. A kidney-shaped area is faintly defined on 

 the neural valve. A portion of this outline indicates the point of attachment of the 

 perivisceral wall. 



In fig. 24a an enlarged view of the digestive sac and adjacent parts of fig. 24 is shown. 

 The liver in this stage communicates with the stomach by large openings on each side, and 

 fine granules were seen rapidly circulating to and fro, from the liver to the stomach. 

 The fit'ces, rolled into a spii'al and pyriform shape, were constiintly in motion by the action 

 of the cilia lining the stomach. This mass was frequently urged toward the mouth 

 before it was finally discharged. The passage of the fteces through the mouth was repeat- 

 edly witnessed ; though after the careful investigations of Huxley, Hancock. Lacaze- 

 Duthiers and Gratiolet, no further proof is needed of the absence of an anal outlet to the 

 intestine of this, and allied species. 



The experiments of Mr. Hancock, such as bursting the intestines mider a compressor, 



MEWOICS BOST. SOC. NAT. HIST. VOL. II. 9 



