MOTHS FROM SOMALILAND. 93 



With few exceptions the specimens were captured at Manclera, 

 and this place is to be understood when no locality is mentioned 

 in the text. Mandera is, however, always quoted for those com- 

 paratively few species which were also taken elsewhere. The 

 specimens from Hargaisa dated Oct. 1908 were taken during 

 a period of about two or three weeks by Captain Jorkinson. 



Mr. Feather writes concerning the method of capture : — 

 " I may say that all the moths were collected at li^ht. I only 

 remember taking one species — I think a Geometer — otherwise, 

 and that I got in a porcupine-burrow along with a Skipper. I 

 sugared many times, but the only insects that came were ants, 

 and they completely covered the sugar." 



Mr, L. B. Prout, in the introduction to the Geometridse 

 (p. 142), draws attention to the remarkable preponderance of 

 females ; and the same unusual condition is to be found again 

 and again throughout the rest of the collection. Observations 

 made Sept. 12-15, 1915, in Bombay Harbour, on the return 

 from the visit of the British Association to Australia, have 

 led me to believe that nocturnal flights of female Lepidoptera 

 tend to occur during wet weather. The Bombay species in- 

 cluded the females of certain butterflies which flew at night 

 and came to light with the moths. The fact seems to be very 

 interesting and well worthy of a separate communication dealing 

 with these Somaliland moths and my own experience in Bombay. 

 In order to test the relationship of female preponderance to 

 wet weather, I have asked Mr. Feather to supply a record 

 of the rainfall and temperature at Mandera. He kindly replied 

 as follows : — 



" I am enclosing the record of rainfall for Mandera, and 

 have used mvich the same words as I wrote in my diary. My 

 impressions of the rainfall were guided by what I had been used 

 to in England, as this was my first visit to Africa. 



" I cannot give you the amount of rain, as we had no rain- 

 gauge. I should say the annual rainfall at Mandera is about 

 10-12 inches. Wet nights are a great rarity, the rain oftenest 

 coming in shoi't heavy showers in the morning or aftei'noon. 



" The river I mention is, of course, a dry river-bed, except just 

 after rain. The longest time that I remember water running 

 down the channel was for about 8 hours. The river-bed is 

 about 120 yards wide, and in one part was a very small stream 

 of permanent water, which appeared between some rocks and 

 ran on the surface for 20-30 yards before disappearing in the 

 sand." 



