410 MR. J. B. GATENBY ON THE BIONOMICS, EMBRYOLOGY, 



o-uarded and temporarily conserved for the benefit of the parasite. In this 

 we but find the usual relationship of host and parasite, which holds good 

 among all animals. 



0£ the thirty or so parasites inside the body of the PortJiesia similis cater- 

 pillar, not one attacks the gut or body-wall of the latter till the correct 

 time has come ; the parasites then are free to kill their host without at the 

 same time being " hoist with their own petard.^' 



Equally important is the question of excretion and defecation ; in Micro- 

 qaster I have shown the absence of the usual Malpighian tubules, and the 

 presence of a pair of coiled tubes in the terminal vesicle. 



With the exception of the fact of Hymenopterous polyembryony, which 

 amono- other insects is unparalleled, it seems that the condition of the 

 Ifirval Microgasteridse is the most highly specialized larval modification for 

 an entomophagous life to be found in the Hymenoptera Parasitica. 



The manner in which parasitism arose in the Hymenoptera in several 

 difi"erent assemblages of forms is difficult to understand ; the same applies to 

 the parasitic Diptera, such as the Tachinidse. Obviously the parasitic larva 

 needs such a highly attuned and specialized system, that it is very difficult 

 to understand how such modifications lor the modus vivendi could have 

 grown up. Moreover, the instincts surrounding the act of oviposition by 

 one of these parasites are very wonderful. The instinct, which enables the 

 newly emerged mother niuscid to seek out meat or dung for its future 

 offspring, has long been considered one of the noteworthy facts in bionomics 

 of insects. How much more wonderful are the instincts which lead the 

 newly-born adult Microgaster to attack, overcome and oviposit in the Porthesia 

 larva it must find, that induce the Chalcid hyperparasite to seek out the 

 dried bodies of Aphids which have been parasitized by an Aphidiiis ! 



The hyperparasite Mesocliorus has a still more difficult task before it ; not 

 only have the Porthesia similis larvse to be found, but one which has been 

 parasitized must be detected, and then the larvse inside the Porthesia cater- 

 pillar's body must be located and successfully pricked. 



I have no evidence at present with regard to the question of defecation in 

 later staoes in Microgaster or Aphidius. Probably, at the time when the life 

 of the Aphid or caterpillar is no longer necessary, the parasite voids the 

 rectum. It should be noticed that in both the silken cocoon inhabited by 

 parasite and by hyperparasite (PI. 24. fig. 1 & PI. 25. fig. 8) at F there is 

 th(i fsecal pellet, as is usual. The vent possibly opens just at the time the 

 larvae are forcing their way outwards. Mr. A. H. Hamm informs me that 

 there is a free-living fossorial \v'ds,i^,Pse7iulus, which seems to feed up without 

 defecating till full-grown ; this probably happens in the dipterous genus 

 Cacoxenus, and the habit is not therefore found only in true internal entomo- 

 phaoous parasites. So far as I know Microgaster larvse do not regurgitate 

 the waste food matter through their mouth. 



