370 Alexander Petrunkevitch, 



cases the behavior of the male did not show the slightest Variation. 

 Tlie necessary conditions appear to be: right time of year, plenty 

 of food and drink, comparatively warm weather, presence of mature 

 sperm in the spermducts and absence of direct snnlight. In twelve 

 cases the web was constructed in the morning-, in three, in the 

 afternoon. None of the males which canie under my Observation 

 constructed a web at night. Judging from the identity of the actions 

 of the flve males in fifteen separate instances as well as from the 

 analogy with the actions of other arthropods in captivity, keeping 

 in mind tliat instincts are as a rule subject to small variations^ 

 we have to assnme that the male Dugesiella constrncts its spermweb 

 in natnre also, by day. The whole Performance requires some three 

 or four hours for its completion, during which the male is exposed 

 to great dangers since the web must be constructed outside of the 

 hole. The difiference in temperature of the room by day and by 

 night was almost nothing in eight cases while in the other seven, 

 it was very great, but since an instinct once fixed by nature, is not 

 subject to change, we may suppose that the construction of the web 

 by day, is in some way dependent upon the higher temperature of 

 the atmosphere then, in the natural habitat of the tarantula. 



A distinct restlessnes of the male precedes the construction of 

 the web. He walks about in his box and if the lid is not on, tries 

 to get out. Finally he choses a place and begins to weave. The 

 shape of the box has no elfect upon the web. In round jars as 

 well as in Square boxes, in boxes containing earth, dry branches 

 and high plants and in glass boxes with nothing on the bottom, the 

 web is always constructed in the same manner and has the same 

 shape. All the tarantula requires is a wall from which to spiu the 

 web in the form of a sheet which is fastened on the other end to 

 the ground. The height of the wall-end of the web above the 

 ground is dependent upon the size of the spider. While attaching 

 the threads to the wall the male Stands on his front feet (Plate 10, 

 Fig. 3), at times quite straight head downward and spinnerets up. 

 He fastens their other ends to the ground, drawing each thread 

 taut. The width of the sheet is about equal to the length of the 

 body of the male. Besides the longitudinal threads the male spins 

 threads in all directions so that the structure of the sheet much 

 resembles that of the web of an Agelenid. In some instances the 

 middle is strengthened by extra threads as we see it on Plate 11 ^ 

 Fig. 8. The male's next step I call the testing of the web. He 



