378 TBANSA4 riONS LIVERPOOL BIOLOGICAL SOCIETY. 



from the roof to the floor. This cavity was rather larger 

 than some others measured. The nest usually measured 

 from 4| to 5 iuches across. I never saw any excavated 

 material at the mouth of the burrows. Not unfrequentlv 

 some fragments of dry grass, exhaling the petrel odour, 

 were found near the mouth of the burrow, which were 

 evidently portions of nesting material dropped by the 

 bird near the entrance. When this was observed the 

 burrow was always found to be tenanted. Dixon (7) 

 long ago made the same observation. That the same 

 burrow is frequently occupied in successive years is 

 proved by my finding in one of the nests a fully developed 

 but dried uj) embryo of a previous year, will) fragments 

 of shell still adhering, which from some cause had been 

 unable to hatch out. The majority of the eggs taken 

 on Boreray on -June 5th were fresh, but some few were 

 distinctly incubated, whilst I took a perfectly fresh egg 

 on Dun on June 12th; so that it would appear that the 

 laying season is somewhat extended. The native's are 

 unanimous in stating that when the egg is taken the bird 

 does not lay again that season. When the burrows were 

 opened out the birds always retreated to the Farthest 

 corner of the nesting cavity, so as to be as much in the 

 dark as possible, and when the nesting cavity had two 

 exits, which it often had, the bird usually retreated 

 along the one which had not been opened out, but, in no 

 case did it spontaneously fly off. When the birds were 

 taken in the hand some of them ejected a little oil from 

 their mouths, but the majority did not do this. When 

 placed on the hillside with their heads facing up the hill, 

 the birds invariably turned round so as to face down hill, 

 and then after a few steps downwards rose on the wing 

 without difficulty. After a few moments of fluttering 

 hesitation, they flew oft' with a rapid zig-zag Might, with 



