penultimate pseudo-segment, placed veiy slightly more dorsall)' than A, 

 there is a small twin-tubercle (B), each one of the twins terminating in 

 a cluster of 4-5 short curved spines. On the same pseudo-segment, just 

 above each twin tvibercle and consequentlv rather more dorsal in position, 

 is another tubercle (C), usually larger than B but never so large as A. 



In the smaller })ro])ortion of larvae from the Ta})low bulbs the 

 tubercles (C) are brownish in colour, the microscopic bristles forming 

 the skin armature rather shorter, and the minute tubercles of the trans- 

 verse rows on each segment less conspicuous. These larvae are probably 

 those of fuherculatus, the others with tubercles (C) unicolorous and with 

 the skin armature slightly more pronounced being strigatus. This sup- 

 position is contirmed by an examination of the larvae from the Evesham 

 parsnips, though a complication arises by the fact that among these larvae 

 are to be found a few which are most obvioush' distinct ; they are whiter 

 in colour, tubercles (C) absent, twin-tubercles (B) rather largei", the skin 

 armature soft, hair-like, and white, and the transverse rows of tiny bristle- 

 bearing tubercles practically indistinguishable. These are, without doubt, 

 the larvae which led Fryer to believe that he was dealing with a distinct 

 species. Unfortunately no insect was reared from the parsnip material 

 to which these larvae could be referred, and they therefore remain at 

 present a myster3\ 



EXPLANATIOX OF PLATE III. 



Tlie genitalia are shown after baviHg been dissected in tlie following 

 manner: — The aedeagus (tigs. 6 and 9) has been removed, and the remaining 

 "shell" bearing the pubescent anal lamellae and the large side-lamellae or 

 "chispers" has been split iulo two corresponding and identical halves, one 

 of which is represented in tigs. 5 and 8 as viewed from the inner side. 



lu its natural position the aedeagus is connected to the "shell" in the 

 following manner : — The upper margin of its basal part (indicated by the upper 

 leit-hand corner of the profile figures) is connected to a plate,* which appears to 

 be a continuation of the inner skin of the side lamellae, and which separates the 

 closed space wherein lies the rectuui from the open space between the side- 

 lamellae where the aedeagus has free play ; in addition, the base of the aedeagus 

 is connected on each side (at a point about half-way down from the point of 

 connection mentioned above) with that part of the outer skin of the " sliell "■ 

 shown as the lowest point in figs. 5 and 8. As a consequence of these attach- 

 ments the end of the aedeagus comes about opposite to the bristle-bearing pads 

 on the inner side of the lamellae. 



The end of the aedeagus in naturally split (in the plane of the paper of the 

 plate) for some considerable distance downwards into two exactly similar 

 halves, the end of tlie penis lying between these two halves. Two curved 



*TIiis " plate " is shown in the figures {in a somewhat unnatural condition owing to dissection), 

 as a more or less pointed prujectiou running diagonally tu the left-hand lower co.'uer. 



