I04 Journal of Comparative Neurology. 



times high and sometimes low. Examples number VII, XXI 

 and the following example are illustrations of a modification 

 that is quite common. 



Example XXV. — Location : angle between brick-wall and 

 water-pipe. Main sheet triangular, attached in the angle. In- 

 clined, lowest where it joines the pipe. Guard web highest at 

 the pipe, gradually diminishing in height as it approaches the 

 brick wall. The following is an extreme case of the same sort. 



Example XXVI. — Location: angle between the book case 

 and and the wall at level of the top of wainscoting. Main 

 sheet inclined sloping at an angle of 45 degrees from the top of 

 the wainscoting up to the book case. Guard web present only 

 at the' lower extremity of the web. There it forms a low fence 

 along the outside of the wainscoting. Snares few. Here again 

 we have an incline up which insects may be tempted. 



The arrangement of the snares is so irregular that I will pass 

 them by with a few words. These snares are sometimes absent; 

 at other times they very from a few threads to a dense mesh- 

 work. Occasionally the arrangement is peculiar. 



Example XXVII. — Location: pile of vertical and hori- 

 zontal logs. Main sheet irregular, nearly horizontal, arranged 

 so that insects could pass from the top the logs on to the web. 

 Gallery in one cornei. Guard web absent. Snares abundant, 

 at a short distance above the main sheet these threads from a 

 horizontal network. 



The secondary sheet cited in example XIII may be but an 

 extreme of such a horizontal net work of snares. 



A number of experiments have been made to test the spider's 

 ability to vary the arrangement of its web to suit the environ- 

 ment. These experiment consisted in taking spiders from var- 

 ions localities and subjecting them to a different environment. 



Example XXVIII. — A large number of spiders were 

 placed in cylindrical bottles and left two days. Each bottle was 

 closed with a perforated cork and only one spider was placed in 

 a bottle. When captured some of these spiders were occupying 

 triangular webs, some were occupying rectangular, and some 

 were occupying irregular webs; but none were occupying circu- 

 lar webs. The majority of the spiders constructed webs. All 

 the webs were circular. As a rule the webs were constructed 



