OBSERVATIONS ON THE OYSTEBCATCHER. 49 



retraction of the foot and the syphons. The bird forces its bill 

 well in between the valves by a series of powerful downward 

 thrusts which merge imperceptibly into as powerful upward 

 jerks, for the purpose of extracting the shell from its home. The 

 shell is then carried to a suitable crevice in the rock, on which 

 it is deposited hinge downwards, and where it is opened up by 

 firmly applied lateral leverage. The mollusc is torn out in 

 successive portions, adherent fragments of shell, if any, being 

 vigorously shaken off before each mouthful is swallowed. The 

 greater number of the shells is detached apparently for the same 

 reason as that applying to Modiola, and, owing to the form of 

 the region of the hinge, a support to the shell is essential before 

 leverage can be applied. The bird may spend some time in un- 

 certain wandering in search of such a place, and once a suitable 

 crevice is found it may be used repeatedly by nesting the shells. 

 Usually a depression in the rock is utilized, but an irregularity 

 in the concretionary matter adhering to the rock or a bristly 

 tuft of vegetation will serve. I found a shell resting between a 

 Limpet and a low ridge of rock, and another placed endwise 

 against the same ridge. 



Of one hundred and six shells, the largest measures one'and 

 an eighth inches in length by one and a half inches in breadth 

 (1| in. x lg- in.), and the smallest three-eighths of an inch by 

 five-eighths of an inch (f in. x f in.), the average being eleven- 

 sixteenths by fifteen-sixteenths (^t in. x if in.) ; 77 per cent, 

 are entire and 23 per cent, are fractured. The lines of fracture 

 are very various, and bear no apparent relation to the size of the 

 shell. Perhaps the commonest form is a transverse fracture 

 about the equator of one valve, with comminution of the ventral 

 portion. But vertical and oblique fractures are not infrequent, 

 removal of either the anterior or the posterior end of one valve 

 sometimes occurs, and a general crushing of the valve is by no 

 means rare. Owing to the comparative hardness of the shell 

 and the absence of epidermis there are seldom marks to indicate 

 the mode of entry. I have found abrasions at the middle of the 

 ventral border in a number of cases, and in one a V-shaped nick 

 had been made in the truncated end of one valve. Judged by 

 direct observation, the usual mode of entry is through the 

 ventral border. 



