The Bionomics of South African Insects. 511 



power of squirting a strongly acid liquid from behind. In. 

 the case of the largest species, such as Anthia, the great 

 strength and large quantity of this liquid render it a very 

 efficient protection ; but in the smaller species my experi- 

 ments and observations lead me to suppose that this is not 

 the case, but that the utility of the secretion lies rather in 

 the fact that it enables the insects momentarily to dis- 

 concert their enemies, and this, owing to their great 

 activity, gives them an opportunity to escape. Under 

 these circumstances it is evidently of importance that an 

 attack from an enemy should be directed to the anal 

 portion of the body in order to ensure its receiving the 

 discharge. The anal white patch is especially noticeable 

 in such genera as Polyhirma (semisuturata, henncttii [see 

 Appendix, pp. 547, 548], notata, rufata, macilcnta, etc.) 

 and Piezia {marshalli and mashuna) ; it also occurs in the 

 Cicindelid 3Iyrmecoptera ])olyhirmoidcs w\\\c\\ consorts with 

 many of the above species. It is probable that the two 

 white or yellow spots which are found in so many Cara- 

 hidiB and Cicindelidx. (see Plate XVII, figs. 7-11, 14-19) 

 towards the apex of the elytra have also a similar 

 significance, that of the Cicindelida} being of course 

 mimetic (of Mutillida} and Garabidm)." 



B. Mtttilloid Colcoptera : CUridm, CaraMdte, and Cicindelidx : 

 Primary and Secondary Syna/posematic a7id Pseudaposc- 

 matic Associations. (E. B. P.) 



When forwarding the following mimics of Mutillidm, 

 Mr. Marshall also sent a record of habits : — 



"Salisbury, Jan. 8, 1899. — The Mutilloid type is not 

 uncommon among Garahidaj, Eccojjtoiitcra cufricollis being 

 a beautiful instance of mimicry. But the markings are 

 most developed among the CUridx, of which I have no 

 less than twelve species all exhibiting the Mutilloid 

 marking more or less distinctly; I expect they will be 

 found to be parasitic on MutiUid^e. I have also caught a 

 little spider which is an excellent Mutilloid mimic." 



This species of Mutilloid spider and another sent later 

 have been described by the Rev. O. Pickard-Cambridge, 

 F.R.S., as Prosthesiiiia alhomaculata and Titus lugens. (Proc. 

 Zool. Soc. Lond., 1901, p. 11, figs. 2, 3, Plate V.) 



"Salisburi/, Feb. 12, 1899.— The Mutilloid group is 

 interesting though not so satisfactory from a cabinet point 



