voL. x1.] ROSEATE AND SANDWICH TERNS. 1238 
invisible from any direction, as she sat brooding far in beneath 
an overhanging sod. Further on, in a rather similar position, 
two other pairs had eggs. These six clutches were con- 
siderably incubated—in fact, two were already ‘“‘ chipped.” 
At an entirely different part of the ground, but again on the 
outskirts away from the general din, two more little parties 
of Roseates had begun to lay. Eventually in these two 
colonies, some thirty yards apart, eight and nine clutches 
respectively were laid. 
In one the eggs were beside some flowering plants, the 
sprays of which arched over the sitting birds, and a particular 
egg was so deeply infringed that I passed it by several times 
and only located it at last when the owner “‘ran in.” In 
the other, the Roseates laid in deep herbage, a favourite site 
being between two tufts of grass, when the blades meeting 
above left the bird’s head exposed at one end and the tail 
at the other. Some eggs were in little bowers approached 
by arun. Frequently, to avoid disturbing the canopy, the 
owners would creep in with neck and breast almost touching 
the ground. Although breeding thus in batches each pair 
was somewhat exclusive, and in most cases the nests were 
at least two and frequently from four to six feet apart. Al- 
most all the Roseate Terns showed a marked preference for 
the extreme outside edge, the limit of their territory being 
reached only when the ground descended abruptly to the 
shingle. 
Apparently the desire to conceal their eggs and themselves 
when sitting was due to the intense hostility of the Common 
and Arctic Terns. During the days I spent at this haunt 
the Roseates were buffeted about continually. While high 
in the air they were comparatively unmolested, but the 
moment they swung down or attempted to alight, the others, 
principally Arctics, attacked them viciously. The Sandwich 
Tern, on the other hand, never interfered ; in fact two pairs 
of Roseates had eggs right up alongside a Sandwich colony. 
Once when an Arctic Tern settled near a Roseate’s egg, the 
owner, which was off at the time, immediately swept in and 
drove the intruder away helter skelter. Otherwise these 
beautiful birds showed no fighting propensities. 
The eggs were generally long and pointed—an extreme 
exception was remarkably rounded—with the ground colour 
ranging from light stone to warm buff, freely speckled with 
rich and dark brown. The underlying markings were of 
blue-grey. High up on some eggs blotches occurred and, 
together with large spots, occasionally formed handsome 
wreaths. The eggs were so characteristic that I was able 
