1S4 Mr. 11. Scott OH 



an exceptional dronglit had prevailed for some time. l\v 

 elinihing on to the ledges, aeeess was obtained to the smooth 

 elifl-faees, wliieh were covered in phices witli jjreat numbers 

 of the hirva^ these appearing (hu U against the grey baek- 

 jiround of rock. As has been found by previous observers, 

 the larvie, when disturbed, sliift tlieir i)osiiion slightly with- 

 out becoming detached from the rock. Considtriil)!c dilhculty 

 was experienced in collecting tlicm, because, when detached, 

 tliey Mere so easily and rapidly swept away by the swift- 

 ialllng water. I did not then know of the method of col- 

 lecting such larvae employed by Dr. Lutz in Hrazil. He 

 states [op. cit. p. 8:2) that if they are covered with the hand 

 and rolled slightly to and fro, they loosen themselves from 

 the rock and fix thcmj-elves firmly by their suckers to the 

 collector's hand ; tluis a number may be collected in a sliort 

 time. Stdl greater difficulty was experienced with the pupai : 

 these are cemented to the rock by their adhesive pads, and 

 when force was used they broke loose suddenly and were not 

 easily caught before being swept away. In such a situation 

 as this, it is not hard to inuigine how rapid the unfolding of 

 the M'ings must needs be on the emergence of the imagines, 

 if any of the latter are to avoid being caught by the rushing 

 Mater — a rapidity of unfolding provided for, as is well known, 

 by the wings being already developed to their full size in 

 the pu|)a, a condition which necessitates a complex system 

 of folding, which in its turn produces the "secondary 

 veining'^ so characteristic of the family. 



Besides the Bl. pharocerid larvae there were on the steep 

 rock-faces some Simuliid larvaj, also curving tubes composed 

 of web and grains of sand, inhabited by larvae of a Psycho- 

 myid caddis-fly. Small caddis-flies were flitting in shady 

 j)laces over the surface of the falling water, and occasionally 

 settling on dry rock close thereto. Some of these wez'e 

 ca[)tured, and have been described by G. Ulmcr"^ as Melano- 

 trichia insulai'is, a new spee.es of Psychomyid ; they are 

 considered by him to be the imagines of the tube-inhabiting 

 larviC. 



The vegetation bordering the stream below the fall was 

 swept in the hope of obtaining imagines of the Blephai'o- 

 cerid. As has been said, this hope Mas not then fulfilled, 

 but specimens were obtained of a fly of such leniarkable 

 appearance that at first sight it was thought to be the special 

 object of search. It proved, however, to be a Mycetophilid 

 of the curious genus Lyyisturrhiaa { = ProbuIaiUs), and has 



* Di'iU.^cli'j eiit. Zeit^clir. V.)l-i, p. .>6. 



