NOVTTATES ZOOLOaiCAE XXVI. 1919. 311 



very oily looking, but deep cadmium rather than ochreous white, the dark 

 portions being glossed with blue. In the final instar whiter than thoas, the 

 dark parts being olive-green instead of brown, with delicate touches of blue. 



Androgens generally grows to larger size, and always possesses a white patch 

 in the form of a little fish set in the brown about the middle of each side. The 

 pupa is longer and even more like dead wood, the thoracic hump being an 

 enormously projected cowl in comparison with thoas. It is often adorned 

 with a touch of green, simulating lichenous growth on dead wood. 



Over a limited period both ova and larvae were freely obtained from the 

 Souza and Utinga districts, and a fine series of the butterfly in both sexes was bred. 

 The female was several times seen on bright mornings in the act of ovipositing. 

 A friend who reared some of the larvae excelled my good fortune by producing 

 three females with large patches of bright yellow in the bronze-green of the 

 forewing, whereas my own specimens were only dusted with yellow scales. 

 Since then, however, I have taken this form exclusively at Manaos, Porto Velho, 

 and Iquitos, the species appearing to be commoner up-river. 



Anchisiades Group. 



Passing now to the next group, we reach in P. hyppason a very remarkable 

 species. I am doubtful as to whether it is rightly placed in this group at all, 

 for I can see no close features which it shares in common with anchisiades, and 

 a good many, so far as its early stages are concerned, that connect it with the 

 Thoas group. In fact, I would put it back over the fence, or give it the honour 

 01 a Hyppason' group all to itself. As the early stages of hyppason were hitherto 

 unknown, and I am now fully acquainted with them, perhaps I may be pardoned 

 for stating my opinion. True, the butterfly of hyppasori bears no outward 

 resemblance to the yellow-and-black-tailed PapUios which we have so recently 

 been discussing, and from its general mimicry of such a butterfly as lysander 

 any novice might be pardoned for placing it in the Aristolochia Division. In 

 this, however, he would undoubtedly be wrong, for as sure as it is a Papilio at 

 all it belongs to Division II. My arguments for placing it nearer to thoas than, 

 to anchisiades are as follows : 1. The egg is large, deep yellow, and deposited 

 singly. 2. Its food-plant, in Para at any rate, appears to be exclusively Piper 

 helemense. 3. Its larva, unlUte anchisiades v/liich is brown and lives in large 

 gregarious batches at the base of orange trees, etc., is of the " bird's dung " 

 type, and in natural position as well as in colour and design it more reminds 

 une of thoas and androgens. 4. Its pupa similarly, differing from the squat and 

 particoloured character of anchisiades, is possessed of well-developed " ears " 

 and thoracic hump, again approximating more closely to the thoas type. Un- 

 fortunately, I omitted to figure the pupa of anchisiades, but it may be said to 

 resemble a heavy edition of torqiiatus without the front projections. Compare 

 the figures of larvae on Plate II and of pupae on Plate IV. 



P. hyppason (pi. ii. fig. 11, pi. iv. fig. 5). 



At certain times of year quite a common species about Para, especially 

 in the larval form in April and May, but not confined to these months. 



Though it is impossible to give the exact times of appearance, I have noticed 

 that a small percentage of pupae " stand over " for several months, while others 



