226 THE SHELDUCK AND SURFACE-FEEDING DUCKS 



various other species with which he commonly consorts. To justify 

 the generally accepted view, the shoveler should feed mainly on 

 minute Crustacea water-fleas, cyclops, cypris, and the larger Gam- 

 marus, and so on. But though large quantities of these small 

 Crustacea may indeed be eaten, no evidence of this fact has so far 

 been brought to light. On the contrary, all who have ventured to 

 describe the menu of the shoveler have enumerated items which are 

 just as eagerly sought, and as successfully caught, by other species 

 not so elaborately lamellated. One author, Seebohm, it is true, says 

 that the shoveler sifts a larger quantity of mud in a given time than 

 any other duck and we may assume he earns more for his pains 

 but this is not enough. So far as the evidence goes, the shoveler 

 would get on just as well with a much less perfectly developed sifting 

 mechanism, and in just so far we seem to be justified in regarding 

 these lamellaB as affording another of the many instances of " hyper- 

 tely " which have of late years been brought together. That is to say, 

 these lamellae have developed beyond what is needed to attain their 

 end. Here, then, is a point well worth the while of the field ornitho- 

 logist to take up. The sequel may prove that the generally accepted 

 views on this theme are correct ; but until the matter has been made 

 the subject of special inquiry, the need for this very elaborate sifting 

 apparatus must remain open to question. 



Having regard to the peculiarities of the beak, it is certainly 

 significant to find that the shoveler in its choice of haunts is a strictly 

 fresh-water bird. It resorts to the sea only when driven by the 

 severest weather ; and it shows a marked preference for bog and reedy 

 marsh-land affording open patches of water, or the backwaters of 

 rivers where there is plenty of cover in the shape of reeds and similar 

 water-plants. This choice is also determined by its small feet, which 

 are ill-adapted to stemming the current of tidal water. Secure in 

 these fastnesses, it seems to spend much of the day in sleep, rousing 

 into activity with the twilight and feeding eagerly till dawn. Night, 

 however, is not the only feeding time, for the shoveler is particularly 



