372 Transactions of tlie Society. 



of liquid before this has dried, and before the meningococci present 

 have had time to " bite " into the solid medium. Flexible collodion 

 makes a good seal, but drying at incubator temperature still takes 

 place in three to five hours, and owing to the use of an oil- 

 immersion lens the employment of a moist chamber does not 

 appear practicable. 



The alternative method of watching growth in liquid culture 

 was to make shallow cells, as suggested by Mr. Martin Duncan, 

 who made several for me of shellac varnish on an ordinary slide,, 

 and to cover with a sterile slip, inserting sufficient medium to 

 ensure flooding of the top of the walls of the cell, and, in con- 

 sequence, automatic sealing thereof by capillary attraction. I 

 now employ cells made with black enamel, which stands steriliza- 

 tion better than the shellac, each cell being about one-tenth of a 

 millimetre in depth. 



By this method relative immobilization of the organism imder 

 observation is secured, and on the whole the method gives satisfac- 

 tory results. 



These results, however,rcquire the greatest care in interpretation, 

 owing to the presence — firstly — of the almost invisible organism 

 referred to, early gemmation and endosporulation being often 

 exactly simulated by minute organisms brought into apposition by 

 slight streaming movements still persisting ; and owing — secondly 

 — to slow rotatory and dipping movements. These movements 

 lead to error, unless careful and prolonged watch is made, becatise 

 an early bud, or a minute endospore, under observation may eitlier 

 temporarily vanish from view, or without having been previously 

 seen may come into view for the first time, and simvdate a fresh 

 bud or endospore. 



In all cases, therefore, most tedious and prolonged observation 

 is necessary, which ought to be repeated again and again with 

 fresh cultures. 



By the use of these cells cultures can be kept liquid for forty- 

 eight hours without difficulty. Disappcintment is, however, 

 frequent, owing to death, or arrest of growth, of an individual 

 oi'ganism under observation. 



Illustrative Fekmentation and Serological Tests of Serum- 

 broth Cultures of the Meningococci employed i^^ the 

 ABOVE (Observations. 



Fermentation Reactions. 



In the case of all the fourteen cultures, details of which are here 

 recorded, no reaction was observed in lactose, saccharose, maunite, or 

 dulcitol. 



In twelve of the cultures acid was produced quite definitely in 



