930 RICKETTSIA 



existence in insects and in animals ; infection of insects occurs in these cases by 

 blood-sucking. There is, however, no evidence to suggest that a separate cycle 

 of development occurs in either host, as is so frequently observed with the protozoa. 

 Morphologically, in the gut of the louse rickettsise appear like very small cocci, 

 diplococci, or short bacilli. Their size is generally given as 0-3-0-5 ju long 

 by 0-3 ju broad, but the bacillary forms may reach a length of 1-5 or even 2-0^. 

 According to Elford and van den Ende (1944) the size of R. prowazeki, judged by 

 ultra-violet photomicrographs, is 0-6-1-8 fj, long by 0-45 ju broad. Gradocol mem- 

 brane experiments likewise suggest that the width is not less than 0-4-0'5 /j, — 

 a value greater than that which has previously been quoted. Their arrange- 

 ment is variable ; single forms may occur ; diploid forms are very common ; 

 and the small coccoid forms are often grouped in dense masses. Under 



certain conditions, they may 



form intracellular homogeneous 



or granular inclusion bodies not 



J^ . \ *='-^ unUke those seen in psittacosis 



J^'^ I - (Pinkerton and Hass 1932, 



* V "A *^^^ microscope they appear, 



y I J ^^^^^ J like bacteria, to possess a limit- 



^\ Hn^^ y^ f^3 i'^g membrane, enclosing a 



^ ' * ' £, ' substance of moderate opacity ; 



} ^l^i^ ^j '^ / dense granules may be seen in 



^J J V ' some forms (Plotz et al. 1943, 



.:^ Weiss 1943). Most species are 



non-motile. On the whole they 



Fig. 222.— Rickettsia prowazeki. stain poorly with the ordinary 



Smear preparation from louse's gut, showing bacillary aniline dyes, but Castaneda 



forms, and occasional thread-like chains. Magnifi- Q93Q) ^^ ^^^g (1932) Laigret 



cation 2,000 diameters (approx.). (After Wolbach, ^ , / , • ,-,\nr^^ t>? 



Todd and Palfrey.) and Auburtm (1938), and Mac- 



chiavello (1941) have shown 

 that good results may be obtained with methylene blue, thionin, and fuchsin, 

 provided the stain is made up with a suitable buffer or mordant. Macchiavello's 

 stain, in particular, gives excellent results with smears, though it is not so successful 

 for tissue sections. For general purposes, Giemsa is probably the best stain ; 

 with this the rickettsise appear as purple cocco-bacilli, or frequently as bipolar- 

 staining rods ; sometimes they seem to be surrounded by a paler-staining sub- 

 stance. They are uniformly Gram-negative. When very few in number, they 

 cannot be recognized with certainty, since their resemblance to minute particles 

 of detritus is too close ; but when they are present in large numbers, their 

 recognition is comparatively easy. It is by reason of the small numbers in which 

 they occur in the blood of infected men and animals, that their demonstration 

 in this medium has only rarely been accomplished. Arkwright, Bacot and 

 Duncan (1919-20) lay down the following criteria for the recognition of Rickettsia 

 in the louse's gut : (1) its minute size, 0-3-0-5 /j, X 0-3 jU ; (2) its irregularity in 

 shape, round, oval, diplococcal, and bipolar-stained bacillary forms being seen ; 

 (3) its occxirrence in very large numbers, or even masses, especially on flakes 

 of solid material in the excreta ; (4) its well-stained appearance and purple 

 colour with Giemsa. • 



