OBSERVATIONS ON THE OYSTERCATCHER. 45 



bill sidewise. These examples indicate the difficulties the birds 

 sometimes meet with, and they also bring out an interesting 

 point. In the ordinary Mussel shell presenting the dorsal 

 border the two valves meet at an acute angle, and by their 

 mutual interaction offer the maximum resistance to pressure 

 from without. Partly from direct observation and partly by the 

 inspection of the shell-remains, it became clear that the special 

 purpose of the two manoeuvres last described was to rotate the 

 one valve partially off the other, thereby to deprive the one of 

 the other's support, and so make it the more easily crushed. In 

 these two cases the attempt was made to depress the one valve 

 below the level of the other by sheer force in small and compara- 

 tively weak Mussels which were, however, presumably resisting 

 to their utmost. A third disadvantage is brought to light by 

 examining the Mussels after the birds have left. The shells are 

 usually cleaned out in situ, and, in general, it is not common for 

 a shell opened by hammering to be removed to a bare place as a 

 preliminary to the removal of the contents. Most bear evidence 

 of having been opened through the dorsal border, and show the 

 ordinary line of fracture extending from a point on the border 

 posterior to the hinge to the anterior end. When the elliptic 

 portion remaining with the hinge is broken into two portions, 

 the posterior part shows marked depression. In addition, when 

 the two valves are approximated, a hiatus remains in one valve 

 posterior to the hinge, and generally the part corresponding to 

 it is found in fragments inside the shell. In some shells this 

 hiatus alone is present, and shows that levering had not been 

 necessary.* A few show that they were hammered open through 

 the posterior end, and a still smaller number through the ventral 

 cleft. In these last the bill was driven truly between the valves, 

 but the margins show signs of rougher usage than do the mar- 

 gins of shells opened by thrusting, and, as in the majority of 

 these, there is no fracture. The shells are rarely well cleaned, 

 the attachments of the mantle to the valves being generally left 

 untouched, and the birds refuse those portions of the flesh which 

 have got mixed with the fragments of the shell driven into them 



* Cfi ibid.,]). 204, where I have advanced the depression of the posterior 

 fragment and the presence of a hiatus as material evidence of the accidental 

 opening of the larger shells where no leverage was required. 



