G. S. GrRAHAM-SMITH 
21 
between 100 and 500 diameters in the different series, as it was found 
impossible to obtain satisfactory results with the higher magnifications 
with some species when growing on the surface. Growths in the depth 
were necessarily only slightly magnified as the organisms do not remain 
in the same plane. The photographs reproduced in Plates III, IV, Y 
are shown at. the original magnification, but some of those on Plates VI 
and VII have been somewhat enlarged. In all cases the prints were 
obtained from untouched negatives. Very fast plates were generally 
made use of in order to reduce the period of exposure as much as 
possible. If a long exposure is given a certain amount of movement 
invariably occurs and spoils the negative. The exposures usually varied 
between five and ten seconds, though sometimes it was necessary to 
give up to 30, 60 or even 90 seconds. 
As soon as the exposure had been made the microscope was taken 
back to the thermostat and the preparation watched until sufficient 
development had occurred to render another photograph desirable. 
The greatest care had to be exercised in moving and clamping the 
microscope as the slightest jar was apt to completely spoil the specimen, 
and many were ruined by such accidents. 
Owing to the frequent changes in temperature caused by removing 
the microscope from the thermostat to the camera development was 
slow, and in order to obtain a satisfactory series the preparation had to 
be started very early and watched for several hours. 
I wish here to acknowledge my indebtedness to my laboratory 
assistant, Mr J. Charles, for the great help he rendered in the solution 
of some of the difficulties met with in obtaining these photographic 
records, and for the care and attention he bestowed on the negatives. 
General observations. 
In order to satisfactorily follow the various changes which occur 
in the formation of a small colony from an isolated bacillus several 
hours have to be devoted to watching a single preparation. Moreover 
owing to various causes such as interruptions, accidents and unsuccessful 
preparations it was almost invariably necessary to spend several days 
watching a number of preparations of a single species. As only 
occasional days could be devoted to this work the observations have 
extended over a period of more than two years. During this time the 
mode of growth of a considerable number of species of bacilli and 
vibrios has been investigated. 
