ASEXUAL REPRODUCTION IN SAGARTIA 165 



of a pair of mesenteries, the latter a space lying between two 

 adjacent mesenteries not members of the same pair. It will be 

 convenient to distinguish between the endocoels partially enclosed 

 by complete mesenteries and those bounded by incomplete 

 mesenteries. The former I shall call 'complete endocoels,' the 

 latter, 'incomplete endocoels.' Obviously, the spaces between' 

 members of a pair of complete mesenteries may be further dis- 

 tinguished as directive or non-directive endocoels, and incom- 

 plete endocoels may be designated according to the cycles of 

 the incomplete mesenteries bounding them. 



The white bar on the disc extending radially from a siphono- 

 glyph marks the position of a directive endocoel, with which 

 the cavity of a so-called directive tentacle communicates. Or- 

 dinarily the orange stripes seen on the column of the living 

 animal mark all the complete endocoels and the endocoels of 

 the first cycle of incomplete mesenteries (p. 207). In very small 

 specimens these orange stripes may be wanting entirely. In 

 regenerating regions they appear rather late and, so far as has 

 been determined, without regularity (p. 214). In such regions 

 no dependence can be placed upon the number of orange stripes 

 in determining the number of complete mesenteries. By the 

 time the regenerating area, which is at first paler than the older 

 portion, approaches closely to the old region in depth of color, 

 the orange stripes have developed in full number and, almost 

 without exception (p. 213), accurately denote the number of 

 mesenteries of the first two cycles. 



The specimens used were collected at various places and at 

 all seasons. They were kept in the laboratory, for as long a time 

 as desired, in small dishes of sea-water. When first taken to 

 the laboratory the water on them was changed every few days. 

 After algae became abundant on the walls of the dishes, changes 

 were necessary only in order to compensate for loss by evapora- 

 tion. Three or four times a week the slime which accumulates 

 over the surface of the animals was washed off with a pipette. 

 Feeding was attempted by placing minute shreds of fish or other 

 meat upon the tentacles. This was effective, but very laborious. 

 It was also hazardous, since any particles of food left in the 



