6o N. \\\.\K\)M.v.: The Fauna of Brackish Poiiiis. [Vol.. I, 



wall of this region more transparent and makes it possible to distin- 

 guish the position of the mesenteries externally. The zooxanthellee 

 are always more numerous towards the distal end of the endoderm 

 cells than at their base, from which, indeed, they are practically 

 absent. 



The second factor is not very important so far as coloration is 

 concerned. It consists of irregularly shaped solid particles and 

 globular masses of liquid, both very minute, occurring in the cells of 

 the ectoderm of the stomodseum and the endoderm of the mesenterial 

 filaments. Other particles, possibly of an excretor}- nature and of a 

 shining white colour, are present in certain cells of the endoderm 

 of the tentacles, giving rise to transverse bars. I can find no con- 

 firmation of StoHczka's statement that these bars are due to accu- 

 mulations of nematocysts, for nematoc3^sts are equally numerous 

 throughout the ectoderm of the tentacles. When zooxanthellse are 

 absent from an individual, the solid particles and liquid globules 

 in the mesenteries and stomodseum give these organs a faint pinkish 

 tinge during life. There can be little doubt that such intracellular 

 accumulations of matter are direct products of metabolism. 



The third factor is the cause of the purplish colour noted by 

 Stohczka in the mesenteries of the t3'pical form, and equally con- 

 spicuous in some individuals of the new race, but not always pres- 

 ent either in the tjq^ical form or the other. If any part of the 

 endoderm of an individual with purple mesenteries be examined 

 microscopicall}' , it will be seen to contain numerous bodies of a deep 

 violet colour. With the aid of a fairly powerful objective such as 

 Zeiss' apochromatic D these bodies will be seen to vary consider- 

 ably in shape and size and each to be enclosed in a green and 

 apparently structureless capsule, the colour of which does not dis- 

 appear in spirit. An oil-immersion lens is necessary- to throw any 

 light on their structure, and even under the highest powers they 

 are minute. Under favourable conditions each body can, however, 

 be seen to contain a large number of smaller, densely pigmented 

 spherical structures, evidently spores, surrounding a colourless cen- 

 tral core. I have not succeeded in investigating the structure of 

 the spores owing to their minute size and to the fact that their 

 dense pigmentation is extremely stable. The capsule is pear-shaped 

 or subspherical in most of the bodies, but in the largest its outline 

 becomes irregular; in some cases it is no longer intact and the spores 

 are scattered round it. An examination of a considerable number 

 of sections and other preparations has elicited the following facts 

 as regards these violet bodies. 



After the spores have been set free among the cells of the endo- 

 derm, they increase in size, and a small, comparatively clear circular 

 space appears in the middle of each. In the centre of this space is a 

 dot so minute that I have not been able to make out its structure. 

 At first it is difficult to ascertain the nature of the envelope in 

 which each of the spores is enclosed, but after they have increased 

 slightly in size it is possible to see that each lies in a capsule 

 resembling that of the parent but exceedingly delicate and only 



