540 
THOMPSON YATES LABORATORIES REPORT 
far as the first of the day after sunrise is included in the term ' night.' In this 
connexion it is of interest to note that all the artificial gnat infections carried out by 
the American commission were done during the day-time, S. fasciata being the 
species employed.* Besides feeding on man, I have observed it on dogs, on an 
agouti, and a bat. 
The periodicity of feeding activity was more closely observed on this species, 
but I rather believe that a similar intermittence of rest and activity occurs with 
other gnats, ' night ' as well as ' day ' ones. 
One batch of 5. fasciata was put with a number of living butterflies, but 
they did not feed upon them ; on the assumption that night-feeding insects attack 
night-sleeping animals, and vice versa, an experiment with moths would have been 
more apt, but none were available at the time. It may be mentioned that the period 
of greatest activity of S. fasciata corresponds with the time of siesta of the human 
being. 
On one occasion embryo filaria was found in this species ; the source of its 
infection was not found, but the suggestion is that it would have been a diurnal 
worm. 
The eggs are not laid in an adherent raft ; they are practically separate, and 
are deposited close to the edge of the water, if not sometimes actually on the surface 
of the containing basin. At any rate, they readily adhere to this surface sufficiently 
firmly to prevent detachment by a stream of water ; this is probably of use 
where the breeding place (e.g. s a small tin or a roof gutter) is exposed to tropical 
rain. 
The larva is easy to recognize from its curious opaque appearance ; unfortunately 
no description was made. As has been observed by Dr. Finlay, of Habana, as regards 
spontaneous dryingt I have seen almost full-grown larvae appear very shortly after 
putting water in a ' dried ' basin in which development had been occurring. In two 
cases the time of cycle from laying of egg to imago was noted ; in one the first pupa 
appeared on the eighth, and hatched into a male on the tenth day ; in the other, the 
first hatch occurred on the twelfth day ; in both cases larvae continued to be present 
for weeks after the first hatch. (It need hardly be mentioned that the vessel was 
effectively screened against intrusion of other possible egg layers). 
The adult insect varies considerably in size ; this is true of both sexes. When 
sitting on a vertical wall the abdomen is tilted towards the wall, so that the tail 
nearly touches it (i.e., in the contrary sense to the orthodox Anopheles posture). The 
hind legs are separate from one another, and away from the body the tarsus curves 
downwards, but tends to a more or less horizontal position. 
* It is, perhaps, unsafe to dogmatise absolutely that any gnat is purely a 'night' or purely a 'day' biter. 
+ N.B. — The atmospheric humidity in Para is never very far from the saturation point. 
