iA. rr 



:ir\ . much curved, four days, on September 19, with some intestinal part 

 protnidin-.:. fig. 23. Nt-xt inoniing it hud sundered. About a month 

 later our <>I it- jortioii9 had subdivided, so that now the entire specimen 



multiplied into six or seven known portions, besides some whose 

 Jrogn had not Ix-en followed. 



Portions of the original animal suUlivided by a similar process, in 

 dependent ly of their si/e. Tlie parte recede so far from each otlu-r, and 



..niM-riin- line l>ecomes so much attenuated, that it moves, on m\ 

 ing the vessel. Two parts of this kind, neither of them two lines in 

 length, were half an inch asunder, the connecting ligament RO much 

 .i!..>\r the surface of the glass sustaining the parts, that a small instru- 

 ment could be introduced under it, as it waved with any motion, fig. -I. 



The portions thus separating, are sometimes exceedingly minute. 

 not above half a line long. On first beholding them, I was induced to 

 conjecture that they might be nascent Holothuruo of a different species, 

 from what was truly the adult parent. Fig. 25. 



GomUniag the substance of these and other observations, it appears 

 that the spontaneous division above described ensues thus : The c, 

 men remains stationary during some time, on the side of its vessel, 

 when each extremity broadens and flattens beyond ite usual dimensions. 

 This flattening occupies a considerable portion of the body, but dimi- 

 nishes, from the broadest part anterior and posterior, towards the middle 

 of the animal, so that by gradually augmenting to a central point, it ren- 

 ders the subject somewhat like a common sand-glass. In this manner each 

 extremity, that at one end of a cylinder, for comparison, and that at tin- 

 other end of the cylinder is slightly enlarged towards the broadest part, 

 suppose a fourth of the whole, whence it diminishes to the central 

 point, the narrowest part or neck of the sand-glass, where the rupture 

 will take place. At length a slender hair-like ligament connects the 

 parts, which ruptures as these mutually withdraw, and leaves the origi- 

 nal animal in equal halves, or consisting of a greater and lesser portion. 



The duration of the process is very irregular, and apparently depen- 

 dent on circumstances, not only unknown, but beyond any rational con- 

 jecture. Neither does it seem influenced by size or season, or by any 



