COLOURS OF WATEH-MITES. 1233 



The ditch was full of breeding 3-spined Sticklebacks {Gastero- 

 steus acideattts). The wiales were guarding their nests and 

 driving off all intruders fiercely. The mites were not abundant, 

 only about 20 specimens being taken, after a good many days of 

 watching and collecting. Tliis fact made it impossible to do as 

 many experiments as one could ha,ve wished. When a ficarlet 

 mite did appear, it was extremely conspicuous and sometimes 

 swam near a male fish without being touched. 



3, The Evidence for Warning Coloration. 



Experiments were made in May 1921 in order to see vvhether 

 mites are eaten by Sticklebacks. 



Eiejjt. 1. A large male 3-spined Stickleback from Railway 

 Ditch was placed in a vessel of water, and starved for three days. 

 It was then offered a Daphnia pulex, which it devoured. 

 Directly afterwards a Limnesia fulgida was put in. The fish 

 made towards it from an inch or two away, but stopped half or 

 quarter of an inch from it and swerved ofi". It repeated 

 this after a minute or two. After that the mite was disregarded 

 altogether. When given another and brighter L. fulgida the 

 fish went up to it once in the same way, and afterwards ignored 

 it. It was given another JJajy/mia w^hich it devoured eagerly. 



I^xpt. 2. The last experiment was repeated on the same fish, 

 Eylais hamata being used instead of Li7)inesia. The mite was 

 completely ignored after several inspections. 



Expts. 3, 4, 5. The same result was obtained when Diplo- 

 dontus des2nciens, Fiona longipalpns, Hydrarachna disiincta, 

 respectively, were used. 



These experiments show that the fish avoided scarlet mites 

 even when it was fairly hungry, that the latter were presumably 

 distasteful to it, and that' the fish remembered this fact from its 

 former experience. 



Uxpt. 6. A Stickleback was placed in a vessel as before. 

 A Dajyhnia was eaten eagerly. An Eylais hamata was put in 

 and avoided by the fish. Another Lajohnia, was eaten. A 

 Hydrarachna fuscata was now introduced. The E. hamata 

 accidentally bumped into the fish, which whipped round and 

 snapped up the mite, but immediately spat it out again. The 

 fish then approached H. fuscata but did not touch it. A iJaphnia 

 was then put in and eaten. 



This experiment shows that E. hamata was actually distasteful 

 to the Stickleback, and that H. fuscata was avoided. 



Mr. Soar tells me that he has never seen any mite eaten except 

 by other mites. He once observed a Dytiscus larva which refused 

 to touch a Diplodontus despiciens. Piersig [3] quotes an obser- 

 vation of Duges to the eflect that ISfejKo refused to touch water- 

 mites, but does not state what species. Thus, scarlet mites are 

 avoided by two of the fiercest enemies of fresh-water animals, 

 DytiscifiS and Sticklebacks, both of which were present in 



