EMERY: SIMUUUM VITTATUM IN KANSAS. 335 



through that ordeal all right, and two flies emerged from the 

 pupje August 4. However, at this time the Lawrence city 

 water was so impure that they were using strong chemicals 

 to purify it. This seemed to kill the larvae, so that they were 

 all gone from the ripple August 5. No more flies emerged 

 from the pupte. Larvae newly hatched from the eggs in the 

 laboratory ripple soon died. On July 2, I placed SimnUum 

 eggs in the laboratory ripple, and on the 5th and 7th I found 

 newly hatched larvas, but the city water seemed to kill them 

 after hatching; at least they disappeared. 



Our experiments had been so badly interfered with in the 

 laboratory by drugged water, and the several streams in the 

 state where we were conducting experiments went dry during 

 the drouth of the summer, so that our experiments looked 

 dubious. Fortunately I located a good brood of larva? in the 

 Little Arkansas river at Wichita. Knowing that Simidiiim 

 breed in sewer ditches, and desiring to establish a permanent 

 brood near our laboratory, I decided to take a number of the 

 larvae from the stream at Wichita and place them in the sewer 

 exit, in good ripples and a fall, at Lawrence where it empties 

 into the Kaw river. On October 8 I collected several hundred 

 nearly grown larvae, from 4 to 5:30 P. M., transported them in 

 a wet pack off the rocks to Lawrence, where I placed them 

 alive in the sewer at 7:15 A. M. October 9. October 10 a lot of 

 the larvaj had disappeared. October 11 most of the larvae 

 had disappeared. A piece of cloth that I had left in the sewer 

 with larvae on it smelled strongly of kerosene and had a white 

 sediment on it. The rest of the larva soon disappeared. Since 

 that time I have succeeded in keeping a SimnUum larvte alive 

 in a tin box with a wet cloth for more than seventy-two hours. 



Simnlium pupae, when first formed, are a yellowish brown 

 color, later becoming darker as the dirt in the water discolors 

 them and as the imago develops within the case. With the 

 filamentary breathing gills a pupa measures atout one-eighth 

 of an inch long. (PI. XXXIX, fig. 7.) Their gills or respira- 

 tory filaments arise from a common base on each side. In 

 S. vittatnm the base of one gill divides into two, and from each 

 of these arise four branches, these again each dividing into 

 two, making sixteen tracheal filaments for each gill. 



According to observations made by Miss Phillips and re- 

 corded in her thesis, 1890, the spinning of the cocoon of 



