470 JOURNAL OF ECONOMIC ENTOMOLOGY [Vol. 10 



paper, somewhat less than the length of the test tube and filled with 

 water to about one-half to one inch high, which is sufficient to keep the 

 filter paper moist for a number of days. The test tubes used are about 

 seven and three-quarter inches long with a rolled edge. To prevent 

 the larvae from escaping, a piece of cheesecloth is tied to the open end 

 by means of a small rubber band.^ The larvae will often hide between 

 the sheets of the paper or on the inner side of it, but about as often 

 come to rest between the paper and the glass walls of the tube where 

 they remain for days plainly visible in all their activities and where 

 they can even be examined microscopically. Molts cannot be over- 

 looked as the shed skins appear plainly on the white surface. In the 

 same way excrements may be readily observed and taken out with the 

 paper for examination. Earthworms may be given as food with the 

 advantage that they move about in the test tube and are soon found 

 by the larvae, but meat proved an excellent substitute and was readily 

 taken by all the larvae under observation. This is of some importance 

 as earthworms cannot always be had and the feeding problem then 

 becomes difficult. Food should be given every two or three days but 

 the larvae can go without food for a much longer time. 



Hine has stated that a disagreeable odor developing in the breeding 

 jars seemed to be injurious to the larvae. I have found, however, that 

 the larvae did perfectly well even in the presence of highly putrefactive 

 and ill-smeUing matter, and were not affected by the presence in the 

 tubes of dead earthworms, decaying meat, etc. Nothing needs to be 

 feared in this respect for the larvae as long as they have air to breathe. 

 In one case only a larva died, evidently in consequence of an infection, 

 after remaining in contact with a piece of meat for several hours. In 

 another case a pupa of T. lineola was seen in contact with a lump of 

 decaying meat for several days and then moving away from it by 

 itself. The larvae will sometimes drown in the water of the bottom of 

 the test tube, but even when apparently drowned will often revive 

 when placed in air again. 



I notice that M. B. Mitzmain (1913) in his excellent studies on T. 

 striatus, placed larvae in jars with sheets of filter paper which were 

 partly soaked with mud. He also saw the larvae congregating between 

 the filter paper and the glass walls of the jar, and he is the only author 

 who observed several molts in Tabanid larvae. A. Lutz (1910) used 

 damp moss instead of mud and glass vials to render the larvae more 

 visible; and test tubes have, as far as I know, been used by Baldrey 

 (1913) who, however, did not raise many larvae. The method here 

 proposed, if not absolutely new, appears satisfactory enough to be 

 recommended for more general use especially for the close observation 



^ I find it convenient to keep the test tubes on wooden racks holding twelve each. 



