1 1 6 Journal of Agricultural Research voi. iv. No. 3 



Prowazek (12) concluded from his experiments that it is sometimes possible 

 to infect healthy silkworms with polyhedra-free filtrates. Since Escherich 

 and Miyajima (4) and Glaser and Chapman (6) obtained negative results 

 with Berkefeld filtrates and since Prowazek's filtration results (12) were 

 rather indefinite, the experiments of the writer will be discussed at length, 

 for the reason that their success depended entirely upon proper attention 

 to seemingly insignificant details. 



EXPERIMENTATION 



Small caterpillars are unfit for experimentation; so it is necessary to 

 wait until they are in the fourth or fifth stage before they can be used. 

 This delay adds somewhat to the difficulty, because at those periods cater- 

 pillars are rapidly approaching pupation, and consequently the experi- 

 mental period becomes shortened appreciably. Furthermore, since the 

 weather is usually hot at this time, wilt is very prevalent in most places, 

 and many more caterpillars are infected than earlier in the season, when 

 the weather is cooler. All the experimental material was obtained 

 directly from the field, as caterpillars raised in the laboratory are 

 unhealthy and utterly worthless. 



The first requisite for the experiments was to ascertain whether really 

 healthy material could be obtained. As the lightly infested locaUties 

 promised a greater number of healthy caterpillars than those heavily 

 infested, a number of lightly infested localities were selected and collec- 

 tions were made, provided the wilt disease was not in evidence. The 

 caterpillars were placed in new wooden boxes and shipped directly to the 

 laboratory. Sometimes the disease broke out during transit ; the entire 

 lot was then killed and another collection made. If the insects seemed 

 healthy on arrival, they were placed in autoclaved trays and submitted 

 to a rigid physical examination for four or five days. It has been shown 

 by Escherich and Miyajima (3, 4) and Prowazek (11) that the incubation 

 period of the polyhedral diseases depends very intimately on external 

 conditions; in other words, heat, cold, starvation, poor food, etc., will 

 very soon convert a chronic into an acute case. For this reason these 

 caterpillars were divided into lots and treated in a variety of ways: Some 

 were starved, others were fed with leaves soaked in water for 48 hours, 

 while still others were subjected to the direct heat of the sun. If a single 

 caterpillar died of the wilt disease during the course of any of these treat- 

 ments, the entire lot was discarded and a fresh collection made; but if 

 all remained healthy, as often proved to be the case, the animals were 

 thought safe to use. This method of obtaining healthy individuals may 

 have its faults, no doubt, but its effectiveness will at once become apparent 

 when the experiments are carefully examined. 



Pasteboard boxes, 7 inches long by ^}i inches wide by 5 inches deep, 

 with a cheesecloth lid, were used in all infection experiments. Several 



