946 Transactions of the Society. 



perivisceral fluid. This viscous secretion was abundant, but offered 

 nothing of moment, save some small pale corpuscles of various sizes, 

 which on the application of alcohol became finely granular, the 

 fluid remaining transparent. The size of the corpuscles (lymph ?) 

 varied from the smallest up to the 1/7000 in. When made 

 to roll over they were seen to be doubly convex. Only four curved 

 bacilli were seen in over fifty fields. 



On the 1 1th of August, a second experiment was made with 

 two fresh Eristalis put together, and fed from the agar-agar culture, 

 diluted on the sugar. They were watched feeding several times. 

 Two dejections passed about four hours later contained in the 

 granular debris many pollen-grains, some small, short, non-motile 

 blunt rods, a few bacteria of different sizes, and numberless 

 micrococci. No curved rods to be seen. Between seventeen and 

 eighteen hours after feeding on the culture, many micrococci 

 in chains of four, and a few dumpy curved bacilli, single and in 

 minute clusters, were noticed. Six other dejections were passed 

 the same day, and only fourteen curved bacilli were counted in 

 thirty-five fields on the stained slide. On the 13th, the pollen- 

 grains were few in number, the straight rods pretty numerous. 

 Many of the micrococci were surrounded by a pale hyaline outline ; 

 the curved bacilli were more abundant — none motile. The commas 

 were still pretty plentiful on the 14th, but motionless among an 

 abundance of minute fatty-looking granules, soluble in chloroform, 

 and some partly broken pollen-grains, and on one of these was a 

 tiny cluster of fine commas. 



On the 15th seven dejections, small and very adherent, when 

 examined, showed the commas and straight rods to be rare. There 

 were numerous irregular lumps of a brownish semi-transparent 

 resinous-looking substance, not soluble in dilute acetic acid or 

 alcohol. They had the appearance of some forms of uric acid, and 

 as if built up of imperfect crystals with rounded edges. Four 

 dejections were examined on the 16th without any special dif- 

 ference being noted. On the 17th thirteen dejections were examined, 

 the two most recent- looking separately ; they furnished numbers of 

 pale flat crystals with rounded corners. The commas were rare. The 

 twelve other excreta contained the same semi-transparent resinous- 

 looking little masses, and the curved bacilli were few and motion- 

 less, some double with one tiny mass. The dejections of the 18th 

 differed scarcely at all from the last. The two insects were now 

 killed by chloroform vapour, and the perivisceral fluid examined 

 within three-quarters of an hour afterwards. This viscous fluid, 

 unstained, showed various pale corpuscles which alcohol rendered 

 slightly granular, and were little influenced by the methyl stain, 

 but this rendered evident numerous pale motionless rod bacilli, 

 consisting of four or five joints, and found in both insects ; these 



