302 



JOURNAL OF ECONOMIC ENTOMOLOGY 



[Vol. 8 



very few pupae after the larvae disappeared. Yet the fact that we 

 reared a specimen of each from the stray puparia collected shows that 

 both species had been present. These army-worm pupae yielded as 

 many flies as moths, approximately equally divided between A. analis 

 and G. unifasciata. 



These rearings and observations taken together with those of Pro- 

 fessor Met calf in 1908 indicate that W. 4-pustulata is the chief insect 

 parasite of the army-worm in the central part of North Carolina, and 

 that P. claripennis was also an appreciable factor in 1914. A. 

 analis and G. imifasciata both killed many pupae in western North 

 Carohna in 1914. 



Braconid and Ichneumon parasites (Hymenoptera) were noted or 

 reared on only an insignificant percentage of the army-worms. Sev- 

 eral predaceous enemies were noted in the field but in no case enough 

 to be an important factor. We believe that bacterial diseases played 

 a part, but secured no confirmation on this. The largest natural 

 factor noted was the Tachina-fly family. 



The indentifications are by Mr. H. E. Smith of the Bureau of Ento- 

 mology, Washington, D. C, to whom I desire to express thanks. 



RELATIVE SIZE IN FUMIGATION 



By C. W. WOODWORTH 



The measurements now always taken in obtaining the size of fumi- 

 gation tents are the distance over the top and the circumference. 

 Most tables now in use are incorrectly calculated because equal values 

 have been given to both dimensions. 



It will be evident at once that the distance over is changed with a 

 change either in the height or the diameter of the tree and that there- 

 fore it is the more important of the two. 



The possibilit}^ of adding to this dimension an amount dependent 

 on the difference between the two measurements which will indicate 

 the proportionate size of a tent gives us a new and very convenient 

 and accurate method of rating a tree for dosage. 



This we will call the ^'relative size" of the tent. If the proportions 

 of a tent remained constant the distance over could then be taken as 

 the relative size. Since the distance around always equals or exceeds 

 the distance over we may select as standard a tent with these two meas- 

 urements equal. The amount to be added to this to allow for wider 

 and shorter tents is one fifth of the difference between the two measure- 

 ments. Thus a tent 20 x 30 would have a relative size of 22 and should 



