6 OHIO STATE ACADEMY OF SCIENCE. 
The pupz studied are dull vellowish in color, with an en- 
circling row of spines or stiff hairs at the apical third of each 
-abdominal segment. Characters for grouping the various species 
are located in the spiracles, and also in the denticles at the apex 
of the abdomen. 
When the larva is first hatched a certain amount of yolk 
is present in connection with the alimentary canal, and this furn- 
ishes it food for a time. The young larve of all the species 
studied in this stage possess the burrowing habit, and whatever 
soft material, plant or animal, they can reach they at once bur- 
row into. Consequently little opportunity is offered under nat- 
ural conditions for studying their habits and growth. The larve 
of many of the specimens probably remain in the water or in 
the mud very near the edge of the water throughout the existence 
of this stage for their pupze come to the surface of the ground a 
few inches from the edge of the water just before the adults 
issue; and around fresh water ponds I have observed myriads 
of pupa skins of Chrysops with just the anterior ends projecting 
above the surface of the ground. There is a wide range of vari- 
ation in the habits of larva of various species of Tabanus. Some 
are hatched in stagnant pools, some in ripples of streams and somr 
in marshy ground. Some species live in water for a time and 
crawl out into dry ground; consequently one often finds Tabanid 
larve by digging in dry ground along the borders of ponds. 
Larve are easily reared if taken when nearly full grown, and 
appear to be as much at home in moist ground as in water. They 
eat small, soft-bodied invertebrates of many kinds, even their 
own species. We have kept them from fall until the following 
spring in small covered dishes filled with wet earth. Plenty of 
moisture was applied and earthworms from a greenhouse near 
by were given for food. In order to prevent one larva from 
eating another of its species, only one was kept in each dish. In 
a large number of trials only a few have died before reaching 
maturity. : 
Not much can be said regarding the length of time our 
species remain in the larval stage, but from circumstances one 
is led to believe that, in some cases at least, more than one year 
is consumed by this stage. At Sandusky, Ohio, in July, Igo1, 
Tabanus stygius was very common, but in 1902, at the same 
place, only one or two specimens were seen. It may be that one 
can account for this difference on the ground of more than one 
year being required for the completion of the larval stage. 
FIELD HABITS OF THE ADULTS. 
The adults have many peculiar habits which the collector 
should know in order to become proficient in procuring speci- 
mens; and besides, some of these habits have an economic bear- 
ing also. The larval stage is passed in the water or at least in 
