912 



INSECTA. 



like manner that of the pro-thorax, the change 

 which takes place between these two segments 

 during the metamorphoses being chiefly a 

 shortening of the sternal surface of the pro- 

 thorax, or second segment, occasioned by a re- 

 flexion inwards of a portion of the external 

 tegument, to form the articulation as in other 

 insects. If, therefore, we include the jugular 

 pieces of Straus as the analogues of the parap- 

 tera, the external surface of each of the three 

 thoracic segments will be found to consist of 

 twelve primary and readily demonstrable 

 pieces, making in all thirty-six, the number 

 which we believe always enters into the compo- 

 sition of the thorax, as formerly stated by M. 

 Jurine. But it is probable that the parts de- 

 scribed by M. Audouin, and recognised by our- 

 selves, are not identical with those of that 

 author, since the scutum of the meso-thorax in 

 Hymenoptera, as will presently be seen, and as 

 formerly pointed out by Mr. Macleay,* is not 

 only divided in the median line, but its two 

 sides are also separated by a deep longitudinal 

 fissure each into two parts, the outermost of 

 which Mr. Macleay designates parapsides. 

 This division of the scutum of the meso-tho- 

 rax, if constant in the other orders, would raise 

 the number of distinct pieces to thirty-eight. 

 M. Audouin, who adopts the name given to 

 these parts by Mr. Macleay, states that although 

 he was previously well acquainted with this 

 division of the scutum in Hymenoptera, he 

 did not assign names to the pieces because he 

 considered them rather as mere divisions of the 

 scutum itself than as distinct parts of the ske- 

 leton. The existence of these pieces, therefore, 

 in Hymenoptera, is a circumstance connected 

 with the number and identification of the nor- 

 mal parts of the skeleton, which, it must be 

 acknowledged, offers not a little difficulty, be- 

 cause if it be ultimately found that these, which 

 are so distinct in some genera, the Chrysidida, be 

 in reality normal parts of the mesothorax which 

 are thus shown to exist as such only by this 

 segment being developed to its maximum ex- 

 tent in this Order, it must be admitted that 

 they also exist primarily in all other insects, 

 not merely in the mesothorax, but in the pro- 

 and metathorax, so that the dorsal surface 

 of each thoracic segment must be regarded as 

 formed not of four but of sixteen parts, the 

 praescutum, scutellum, and post-scutellum 

 being each divided in the same manner as the 

 scutum into four pieces, first by a median line 

 into two halves, and these again divided late- 

 rally into two others. The two middle pieces 

 would then constitute the notum, or dorsal sur- 

 face of each segment, and the two lateral the 

 purapsides. An equal number of parts must 

 then be recognised as entering into the forma- 

 tion of the ventral arch of each segment. Each 

 middle or sternal piece, formed of four con- 

 secutive pieces, analogous to those of the dorsal 



* Zoological Journ. vol. i. 



arch, and divided in the median line, would 

 correspond to the middle series of dorsal pieces, 

 the proper notum, and a like number on each 

 side of the sternum would correspond to the 

 lateral portions of the dorsal arch, the parap- 

 sides. Of these lateral pieces of the ventral 

 arch, three are already known in each segment, 

 as we shall presently find, the parapteron, epi- 

 sternum, and cpimcron. But since there has 

 never yet been actually found even an approxi- 

 mation to this multitude of pieces in the dorsal 

 surface or arch of the thorax, we prefer for 

 the present to follow the views of M. Audouin, 

 and with him to regard the parapsides as only 

 detached portions of the scutum in Hymenop- 

 tera, in which the development of the meso- 

 thorax is carried to its greatest extent. It must 

 be acknowledged, however, that in admitting 

 the parapsides to be only detached portions 01 

 the scutum, and not primary parts, the same 

 thing may be urged with regard to some of 

 those which, according to M. Audouin's views, 

 are believed to be normal structures. But this 

 objection seems to be replied to by the fact 

 that the pieces described by M. Audouin are 

 almost always found to exist most distinctly 

 marked, whether developed to a greater or less 

 extent, in the generality of insects. Thirty- 

 six, therefore, we regard as the number of 

 the distinct external parts of the thorax. Yet 

 even this is more than has been recognized 

 by others who have attended to this subject. 

 Thus Knoch describes only twelve, Chabrier 

 and Burmeister, eighteen ; Kirby and Spence, 

 twenty ; Straus Durckheim, twenty-two ; and 

 Macleay, fifty-four. But whatever be considered 

 the exact number, they are never all distinctly 

 developed in every insect, owing to the causes 

 before explained with reference to the greater 

 developement of some parts than of others, al- 

 though some trace of the existence of the atro- 

 phied pieces usually remains. It is owing, also, 

 to the same causes, that the actual position of 

 the parts becomes altered in different insects, 

 although their relative position continues the 

 same. 



Very much confusion has arisen in the de- 

 scriptions of the parts of the thorax, in conse- 

 quence of various authors applying different 

 names to the same parts in different insects, 

 and also from the uncertainty which, as above 

 shewn, exists in the opinions of authors with 

 regard to the exact number of pieces that enter 

 into the composition of the thorax. In order, 

 therefore, to obviate as much as possible this 

 serious inconvenience and difficulty in recog- 

 nizing the parts, we shall add a table of the 

 names given to them by Audouin, with refe- 

 rences to the delineations of them by that 

 author, and also the synonyms used by other 

 writers. In doing this we shall also adopt 

 Burmeister's very convenient names for the 

 upper and under surface of each thoracic seg- 

 ment, which are equally simple, and distinctive 

 of the parts to which they are applied. 



