114 



PSYCHE. 



[July 1S91 



Lina macroblastic. Apart from its being un- 

 scientific to classify things as big and little, 

 it is difficult to understand how such a classi- 

 fication can be of any service whatsoever. 

 There is a complete gradation between long 

 and short germ-bands : in Stenobothrus the 

 germ-band is very small when first outlined 

 on the yolk ; in Mantis and Oecanthus it is 

 somewhat larger; in Gryllus still larger; in 

 Blatta it is nearly as long as the egg; in 

 other forms, like Musca and the Coleoptera, 

 it is longer than the egg. But this is not a 

 difference in the germ-bands, it is a difference 

 in the amount of yolk. Stenobothrus has a 

 direct development, Musca undergoes a pro- 

 found postembryonic metamorphosis; the 

 former needs a great deal of yolk because its 

 embryonic development is long and compli- 

 cated, the latter but relatively little yolk, be- 

 cause its embryonic development is very 

 short and comparatively simple. Although 

 Prof. Graber was aware of the existence of 

 transitional forms between his long and short 

 germ-bands, it seems never to have occurred 

 to him while writing the ten long quarto 

 pages, which he devotes to this and similar 

 distinctions, that the true differences lie in 

 the quantities of yolk with which different 

 eggs are provided. This is a strange omis- 

 sion for an embryologist to make after all 

 that has been said and written on the effects 

 of yolk on development. Verum ofieri longo 

 fas est obrepere somnitm. 



Equally artificial and useless is Prof. Gra- 

 ber's division of germ-bands into straight 

 and crooked (tanyblastic and ankyloblastic). 

 It is obvious that the curvature of a germ- 

 band depends on the character of the yolk 

 surface on which it happens to lie. Thus the 

 germ-band of a spherical egg is necessarily 

 curved (Phryganeidae), while the germ-band 

 on the long side of an elongate, oval egg will 

 be more or less straight (Blattidae). It is 

 somewhat disappointing to find that no at- 

 tention is devoted to the important relations 

 of the germ band to the micropylar axis, a 

 subject on which Hallez has published two 



suggestive little papers (Comptes rendus, v. 

 101, 1SS5 and v. 103, 18S6). 



Prof. Graber finds the abdomen of the em- 

 bryo insect to consist oieleven true segments. 

 He believes that he has found distinct traces 

 of coelomic cavities in the last (eleventh) 

 segment, and figures them in Mantis and 

 Hydrophilus. If correct, this observation is 

 of great interest, since Haase has recently 

 maintained, after an exhaustive study of the 

 facts of larval and imaginal structure, that 

 there are only ten segments in the insect ab- 

 domen, the "afterstiick" not being a true 

 segment. * 



The antenna? are shown by Prof. Graber 

 to be decidedly postoral in their origin. 

 Reichenbach pointed out that of the two 

 pairs of antennae in Astacus the first arises 

 on a level with the mouth, while the second 

 is postoral. As far as their relation to 

 the mouth is concerned, therefore, the anten- 

 nae of insects would correspond to the 

 second pair of antennae in Crustacea. The 

 labrum arises, as Prof. Graber points out, 

 from a pair of appendage-like organs. The 

 honey-bee is cited as an exception to this 

 general rule, the labrum of this species 

 having been described as an unpaired appen- 

 dage from the first. But Carriere has re- 

 cently shown that the labrum of the wall- 

 bee {Chalicodoma muraria') arises as a 

 pair of papillae at first separated at their 

 bases, but subsequently uniting to form a 

 single piece. Prof. Graber has not succeeded 

 in throwing any new light on the obscure 

 question as to whether the labrum represents 

 a pair of true appendages serially homolo- 

 gous with the antennae, mouth-parts, legs, 

 etc. 



A lengthy chapter is devoted to a consid- 

 eration of the abdominal appendages of 

 insect embryos. Among the numerous facts 

 recorded the most valuable are those relating 

 to Hylotoma berberidis. In this Tenthre- 

 dinid the German investigatorsucceeds in es- 

 tablishing direct continuity between the em- 

 bryonic abdominal appendages and the pro- 



