34 MR. K. M. SMITH ON CERTAIN SENSE-ORGANS 



« 



a very thin chitinons membrane which is producefl into the 

 sensory processes much in the same fashion that fingers arise 

 from a glove. These sensory processes vary much in shape; 

 in the particular insect chosen as an example they are bottle- 

 shaped, with the portion resembling the neck of the bottle 

 produced to a great extent. In some cases the abrupt change 

 from the thick surface chitin to the chitin lining the sense- pit is 

 extraordinarily marked, e.g. PI. I. fig. 1. Beneath the floor of 

 the pit is a rounded mass of large radiating cells, each cell sepa- 

 rating from its fellow as it approaches the base of the pit and 

 running to its corresponding sensory process, see PI. IV. fig. 11. 

 The whole mass of cells is embraced by a branch of the large 

 antennal nerve ; this appears black in the photograph, PI. IV. 

 fig. 11, iV^. 



Although, as mentioned above, the view of most workers has 

 been that these large cells are themselves nervous elements, 

 Berlese (1) asserts that this is a mistaken view, and that they 

 are glandular, while the nerve-fibres run between them. He says 

 that they secrete a fluid which fills the sensory processes, comes 

 through the chitin to the exterior, and bathes the whole surface 

 of the pit. He quotes Von Rath as agreeing with this view, 

 also Erichson (1847) and Sulcy (1891) as having affirmed the 

 presence of a " humour " secreted osmotically. Packai-d (4) also 

 states that these sensory processes are filled with a serous fluid 

 and are definitely olfactory. 



(IV.) Chief Modifications of the Sense-Pits. 



As stated above, the sense-pits are modified in form in various 

 ways. The following exainples of this may be mentioned. In 

 the Stratiomyidje (text-figs. 10-12) the processes are in groups, 

 each group arising from a common found;ition, and having much 

 the appearance of a partly closed hand. In the Syiphidae the 

 chief distinguishing feature lies in the large number of sensory 

 processes and the great size of the pit ; also in this family appears 

 a further modification in that there are two kinds of pits, one 

 consisting of a simple inpushing filled with coarse chitinous rods 

 (PI. 1. fig. 2), and with few or none of the thin-walled sensor}?^ 

 processes, the other consisting of the large and beautiful type of 

 ])it figured for many of the Syrphidfe, Erisicdis tenax, Xylota syl- 

 varioni, and others. A somewhat similar difference in the size 

 of the sense-pits occurs in the Muscida?, and here also a further 

 modification is found, namely in the fusion of three or four 

 or sometimes more of the larger pits; e.g. PI. II. fig. 5, in which 

 the pits are double. In some species of the Miiscidas and some 

 Anthomyidse, the entrance to the pit is elongated into a channel 

 whose wall consists of spiral folds or basket-work of chitin 

 (text-fig. 38). The greatest variation occiu's however in the size, 

 shape, and number of the sensory processes themselves, which a 

 glance at the Plates will at once make clear. 



