27Z NATURAL HISTORY I 
LETTER LV. 
TO THE SAME. 
DEAR SIR, Oftober 10, 1781. 
I THINK I have obferved before that much the mofl confiderable 
part of the houfe-martins withdraw from hence about the firfb week 
in O^ober ; but that fome, the latter broods I am now convinced, 
Hnger on till towards the middle of that month : and that at times, 
once perhaps in two or three years, a flight, for one day only, has 
Ihown itfelf in the fiiil week in November. 
Having taken notice, m Otiober 1780, that the lafl flight was 
numerous, amounting perhaps to one hundred and fifty ; and that 
the feafon was foft and ftill ; I was refolved to pay uncommon 
attention to thefe late birds ; to find, if poflible, where they rooft- 
ed, and to determine the precife time of their retreat. The 
mode of life of thefe latter hirundines is very favourable to fuch 
a defign ; for they fpend the whole day in the flickered diftrid, 
between me and the Hanger^ failing about in a placid, eafy 
manner, and feafting on thofe infedts which love to haunt a fpot 
fo fecure from ruffling winds. As my principal objedt was to dif- 
cover the place jpf their roofling, I took care to wait on them 
before they retired to reft, and was much pleafed to find that, 
for feveral evenings together, juft at a quarter paft five in the 
afternoon, they all fcudded away in great hafte towards the fouth- 
eaft, and darted down among the low flirubs above the cottages at 
the end of the hill. This fpot in many refpeds feems to be well 
calculated for their winter refidence : for in many parts it is as 
fl:eep 
