55 



it had retreated for the day, and where, for greater safety, it 

 had gone as far as it could into the upturned portion, care- 

 fully blocking the entrance with debris and soil so as to 

 conceal it. From this subterranean retreat the larvae were 

 with difficulty extracted except by digging them out ; and 

 many suffered themselves, as do worms, to be extracted piece- 

 meal with a pair of small forceps rather than relinquish their 

 hold. Indeed the resistance they offered when occupying the 

 upturned portion of the burrow was extraordinary. As to 

 their numbers, there was an average of 10 larvae to the square 

 foot, or 90 to the square yard ; in the badly infested spots 

 there would be quite double that number. The larvae which 

 were kept under observation were found to leave their burrows 

 at night, returning again to them during the day. Pupation 

 took place throughout the month of August, the first two flies 

 appearing in the breeding-cage on the nth of the same month, 

 and others continued to hatch until the end of September. 

 In the infested area the crane flies were more abundant in 

 the early part of September, when they occurred in myriads ; 

 and on one occasion, during a north-west wind, millions of 

 them were swept together into the sheltered corners of the golf 

 pavilion. 



" During the month of June, when the ' leather jackets ' 

 or larvae were devastating the grass, the Rooks flocked to the 

 infested area and devoured the grubs in great numbers. Their 

 feeding time was during the early hours of the morning and 

 again during the evening, after the golfers had gone away. 

 Small parties were also to be seen at intervals throughout the 

 day. The method adopted by the Rooks for extracting the 

 larvae was simple and, at the same time, extremely interesting. 

 As already stated, each larva retreated during the daytime 

 into the upturned portion of its burrow, where it was perfectly 

 concealed and apparently safe, there being but one exit and 

 that invariably blocked at the entrance, then over all was the 

 additional mat-like surface of grass. Apart from the dead grass 

 there was, therefore, no external evidence of the existence of 

 the larvae. The Rook had first of all to find out the larval 

 retreat, and this it apparently does by probing with its beak 

 and pulling away tufts of grass. Having located a burrow 

 the bird then ascertains the direction in which it is curved, 



