BIRDS MENTIONED BY ARISTOTLE. 245 



metonymy, a claw, in which sense Pliny also uses the word. 

 Further, the Sand-Martin is not common in Greece, as there are 

 few rivers there suitable to its mode of life. It may also be 

 stated that it would be more likely to be seen after heavy summer 

 rains than at any other time, and that its stay in Greece is com- 

 paratively short, Von der Miihle saying that the Sand-Martin 

 migrates south from Greece very early in the autumn, even 

 earlier than the House-Martin. All these facts tend to show that 

 drepanis was the Sand-Martin (Cotile riparia). At one time it 

 seemed to me that the Crag-Martin (C. rupestris) was the bird 

 meant, but, although this is a rather rare bird in Greece, it has 

 been established that it is resident there, according to Linder- 

 meyer, Seebohm, and others. Aristotle's drepanis does not 

 appear to have been resident. Besides, the Crag-Martin seems 

 to be called chelidoni by some modern Greeks, according to Dr. 

 Erhard's ' Fauna der Cycladen.' 



A consideration of the extracts previously given from Aris- 

 totle, and of the structure and modes of life of the Cypselidce and 

 Hirundinidce of Greece, would seem to show that the Swifts 

 [Cypselus apus and C. melba) should be included under the 

 ancient name apous, and that the Swallow (Hirundo rustica), at 

 least, should be included under chelidon. Whether the House- 

 Martin and the Crag-Martin, if that bird were known separately 

 by Aristotle, should also be included under chelidon, is difficult 

 to decide. The inquiry, as far as the Crag-Martin is concerned, 

 may best be abandoned, the data being wholly insufficient. With 

 regard to the House-Martin, Aristotle must have known this 

 bird well, as it is an exceedingly common summer resident in 

 Greece. When the extracts from Aristotle are considered, how- 

 ever, some are found to be not inconsistent with the view that 

 chelidon included the House-Martin ; but, on the other hand, 

 there is one important passage which tends to show that the 

 House-Martin was regarded by Aristotle as one of the apodes. 

 This passage is given in his Hist. Anim. ix. 21, 1, which states 

 definitely that the apodes differ from chelidon in having a rough, 

 hairy, or feathered metatarsus (kvyi/a,yi dao-eTa), a statement clearly 

 applicable to the House -Martin, as Gloger said. Then follows 

 the nesting description, which appears to be intended to refer to 

 the apodes, but applies emphatically to the House-Martin of 



