Dixon — Note on the Walking of some of the Arthropoda. 577 



of the diameters of the shadows will give us the relative weights 

 on the legs, &c. ; and then by weighing the spider we can get the 

 absolute weight on each leg. The diameters of the shadows are 

 easily measured on several of my photographs. In many cases, 

 however, it is to be observed that the spider does not walk normally 

 on the water. 



The photographs of the Tegenaria Derhamii were taken with an 

 exposure of -g^th of a second ; therefore we may conclude — 



(1). That the wave takes longer than the -g-V^h of a second to 

 traverse one set of diagonals ; for within the time of exposure 

 seldom more than two legs, and never more than three, have been 

 raised. 



(2), The motion of one leg is not completed within the -i^ih. of 

 a second, for the complete stroke is never shown in a photograph. 



In these observations the mean rate of advance of the body, 

 which was found by measuring the blur in the photograph, was 

 3 inches per second or somewliat less. 



Further, by measuring on the photographs of running spiders 

 the width of the blur at the extremity of the leg, its linear velo- 

 city at the extremity is found to be about 6 inches per second. 



In the case of the cockroach the exposure was -^-Qih. of a second, 

 therefore — 



1. There was usually more than ^-'o^tli of a second's lag between 

 the raising of the last leg and first leg of one set of diagonals ; for 

 we often find the first and second legs coming to the end of their 

 stroke before the third leg of that set of diagonals is raised. 



2. The stroke of one leg is sometimes nearly accomplished 

 in 4Vth of a second. 



The mean rate of advance of the cockroach measured on 

 photographs was 2*2 inches per second, and the mean linear velo- 

 city of the extremity of the waving antennae 7 inches per second. 

 The mean linear velocity of the extremities of the middle legs was 

 2-7 inches per second. To deduce the angular velocity from the 

 absolute velocity of the extremity of the leg we must take into 

 account, of course, the velocity of body and length of the leg. 



In addition to Professor Haddon and Dr. Wright, I am also 



