FEEDING-HABITS OF THE DUNLIN. 



11 



to admit of section that I saw the nature of the recurved 

 burrows and the operations of the Dunlin upon them. 



The imprints left by the Dunlins on the sand and mud are 

 worthy of consideration. In surface-feeding there is nothing to 

 note save, perhaps, the absence of certain univalves from their 

 tracks. On the areas of open burrows single probings are seen 

 often wide apart, and, as I will explain later, they are of the 

 deep variety. As a rule each coincides in position with a 

 burrow. For an obvious reason, ground under water, very 

 liquid ooze, and wet sand show no markings, or else they are so 

 much run as to be of no value. The firm sand along the high- 

 water mark is best for the purpose. The hidden animals leave 

 no surface-markings, and the Dunlins tap and probe rapidly in 

 search of food. When they have been on this kind of sand for 

 any length of time it becomes covered with the tracks of feet 

 and bill. The imprints made by the bill are of three kinds,* 

 distinguished not so much by the sharpness of their differences 

 as by the frequency with which the average forms occur. They 

 are a slight double dent in the sand made by a gentle pressure 

 with the point of the bill ; a shallow probing, an eighth to a 

 quarter of an inch in depth, usually but not invariably divided 

 into two compartments by a transverse septum of sand ; a deep 

 probing, a quarter to half an inch or more in depth, and com- 

 plete in the sense of having no septum. The relative frequency 

 of the three kinds is variable and depends on a number of con- 

 ditions, of which the appetite of the Dunlin, the nature, position, 

 and relative abundance of the hidden animals seem to be the 

 most important. As much of the sand is covered only at spring 

 tides, imprints are added at each high water during neap tides, 

 until the imprints nearly cover the sand for considerable 

 stretches, especially if the Dunlins are many and no rain has 

 fallen. Excluding sand which has been visited more than once, 

 we find that the distribution of imprints is patchy, crowded 

 together in some places, scanty in other st — that they are more 

 numerous near clumps of seaweed and decaying vegetable matter. 

 The larvae are more plentiful in these situations, and may lie in 

 bundles close to the surface under contiguous imprints, which 



* Macgillivray, ' History of British Birds,' iv. pp. 207-213. 

 t Ibid. 



