4 oo DISCOVERY REPORTS 



In Physalia two sorts of flagellum occur, and they are found in large numbers. They are best seen in 

 gastrozooids fixed while feeding, the buccal endoderm being spread out as a flat sheet (PI. XXVII, fig. 7). 

 They occur in rows. The shorter flagella (about 10 fi in length) occur in groups of about a dozen. The 

 long flagella (about 25 /i in length) are more closely bunched together, and occur in groups of about half 

 a dozen. These numbers are rather variable. The short flagella seem to arise from the centre of the cells, 

 the larger from the sides. Both sorts of flagellum can probably occur in one and the same cell, but this 

 is not always the case, and groups of the longer sort are in a minority. The blepharoplast in both types 

 is single, not double as in Hydra. The endoderm throughout Physalia is flagellated, but the bunches of 

 long flagella are confined to the buccal cells of the gastrozooids. 



1 mm 

 Text-fig. 6. Transverse section through a gastrozooid in the basal region, ec = ectoderm, 

 en = endoderm, mes = mesogloea, v = villi, cut at various angles. 



Turning now to the basal region of the gastrozooid, we find a more complicated organization of the 

 endoderm. The most conspicuous features are the villi. These are finger-like projections from the 

 lining of the zooid, consisting of endoderm cells covering a central core of mesogloea (Text-fig. 6; 

 PI. XXVII, fig. 6); the ectoderm is not involved. The villi can be seen through the transparent 

 wall of the zooid (Text-fig. 4). 



The endodermal cells of the villi are probably all of one type ; they are quite clearly active in intra- 

 cellular digestion. In certain specimens, which had been feeding shortly before fixation, the enteron 

 is filled with a mixture of partly-digested flesh, pigment, both dispersed and in aggregates, and 

 nematocysts, discharged and undischarged. The same objects have been identified in vacuoles in the 

 endoderm cells of the villi, and it may therefore be stated with confidence that these cells engulf whole 

 particles of food, and enclose them in digestive vacuoles. A variety of stages in the breakdown of food 

 can be found amongst the vacuoles. 



The pigment in the food is probably melanin, guanin or some other pigment originating from the 

 dermal chromatophores and iridocytes of the captured fish. Part of it may, however, derive from 

 broken down haemoglobin originating in the blood of the fish. The pigment evidently resists digestion 

 in the vacuoles fairly successfully, for the cells of the villi nearly always contain some, whether 

 feeding has taken place recently or not. Presumably it is eventually ejected, and some of the dispersed 



