1030 PARASITES OF PROTOZOA 



small percentage were they present also in the general cytoplasm, par- 

 ticularly the ectoplasm. Rod-shaped microorganisms were found on the 

 capitulum of Macrotrichomonas pulchra from Kaloternies contracti- 

 cornis (Kirby, 1938a) and have since been observed in that species 

 from other hosts (Fig. 212B). Their variability in number and the 

 fact that they do not occur on all specimens, or on any specimens 

 in certain hosts, are in agreement with the view that they are symbionts 

 rather than structures of the flagellate. 



The greatly expanded capitulum of a large devescovinid flagellate 

 from Neotermes bisidaris is constantly encrusted with short, stout rod- 

 lets which show evidence of fission. In this remarkable flagellate there 

 are similar rodlets in the peripheral cytoplasm, usually separated by a 

 narrow or a broad bacteria-free band from the capitular group. 



Bacteria distributed generally in the endoplasm of termite flagellates 

 are frequent and of many types. Some of them are practically constant 

 in occurrence, as are the slender granule-containing rods in the endo- 

 plasm of Caduceia nova and C. theohromae (Kirby, 1936; 1938a). 

 Devescovinids often contain many deep-staining cytoplasmic granules 

 which may be bacteria. Jirovec (1931b) found two kinds of bacteria 

 very abundant in the cytoplasm of every specimen of Trichonympha 

 serbica he studied. They were not present in T. agiUs, of which T. serbica 

 is probably a synonym, from termites in Spain; so probably they are 

 either present or absent in flagellates of different termite colonies. Para- 

 sites similar to one type (paired cocci) were found by Georgevitch 

 (1929, 1932) in T. serbica. Pierantoni (1936) found minute bacteria 

 present in large numbers in the cytoplasm of Trichonympha minor and 

 T. agilis, and he reported their occurrence to be constant. 



ASSOCIATIONS OF AN OCCASIONAL CHARACTER 



Schizomycetes in Mastigophora. — Bacteria in the cytoplasm of flagel- 

 lates of termites have been reported by Kirby (1924) in Dinenympha 

 fimbriata, by Duboscq and Grasse (1925) in Pyrsonympha vertens, by 

 Connell (1930) in Oxymonas dimorpha, by Connell (1932) in Gi- 

 gantomonas lighti, by Powell (1928) in Pyrsonympha major, and by 

 Kirby (1932a) in Trichonympha. Forms like the crescentic organism 

 described by Connell have been seen by the writer in occasional speci- 

 mens of a number of devescovinids. They are not stages of Sphaerita, 



