NOTES ON DRAGONFLY PARASITKS. 245 



grown, colourless at first, but usually reddish or reddish-brown 

 in later life. Unless they are placed immediately in a preserva- 

 tive fluid, however, they quickly lose both their colour and 

 rotundity. We have always noticed that mites occur most 

 plentifully when their Zygopterid hosts are but newly on the 

 wing, and it would seem that dragonflies rid themselves of their 

 parasites to a great extent as the season advances. We have, 

 however, a mite-attacked male of E. cyathigcricm taken at the 

 Black Pond, Surrey, as late as September 3rd. It would also 

 seem that some larvae leave the egg much sooner than do others, 

 for we have met with teneral specimens of P. mjmphula early in 

 May and teneral examples of E. cyathigenim towards the end of 

 July carrying larvre in the same colourless and ungrown condi- 

 tion. Although, as we have seen, locusts are weakened and even 

 killed by the attacks of parasites, there is no evidence before us 

 of injury done to Odonata in this way. 



The occasional presence of mites upon Odonata was known 

 so long ago as 1778, when De Geer figured and described red 

 parasites from the thorax of a small dragonfly under the name 

 Acai'us lihellulce, and identified them with the A. gymnopteronim 

 of Linnseus. By these and other old writers they were regarded 

 as adult forms, but it was subsequently recognised that these 

 six-legged creatures were larval forms which in later life acquired 

 the fourth pair of legs proper to Arachnida. As we have seen, 

 Krendovsky called some larvae from the wings of an Anisopterid 

 dragonfly Arrhenurus papillator, and, following him, Mr. Soar 

 considered parasites from the bodies of such dragonflies as Agrion 

 pidchellum and A.puella to be no other than Arrhenurus globator. 

 We have made several enquiries with a view to clearing up the 

 doubts which still exist respecting the affinities of dragonfly 

 parasites with other larval Hydrachnidse, but we have failed to 

 elicit any definite information, owing to these parasitic stages 

 being so little understood. Three species of Zygopterides with 

 mites attached to their bodies were submitted for examination 

 to the eminent Hydrachnologist, Herr F. Koenike, of Bremen. 

 He has kindly informed us tbat the parasites belong to different 

 species of Hydrachnidae, but to the same genus. At present he 

 is unable to say with certainty into which genus the larvae fall, 

 but in any case he does not agree with those authors who have 

 referred them to Arrhenurus. Attempts made by Herr Koenike to 

 rear dragonfly parasites have not been successful, but he has bred 

 the real larva of ^. globator, and he is satisfied that the parasites 

 in question have nothing whatever to do with Arrhenurus larvae. 



It was suggested by McLachlan that the parasites may 

 attain their position on the body or wings of a dragonfly by 

 climbing up the legs of their host while it is at rest. This 

 suggestion, in itself a likely one, is rendered the more probable 

 by the fact tbat in our experience of Zygopterides the mites 



