1022.1 245 



sido in rows wliicli may be ciilior st'-aio-lit or in short, spirals. If one of 

 tlie eggs at tlie end of a row is freed l>j a needle-point from the surface 

 of the calamus the otliers lift with it, and hy the exercise of a little care 

 the whole row can be detached, the eggs being glued together so that 

 when turned over the row appears to be a glistening band. The cement 

 extends as a narrow margin beyond the edge of the band (fig. 1 d). 

 I have not observed oviposition by this species, but apparently what 

 happens is that the cement (whose emission takes place at the beginning 

 of the process) runs between the egg-shell and the smooth surface of the 

 barrel, and is not gathered at the posterior (?. e., that remote from the 

 operculum) end of the egg into an anchoring mass. I have, however, 

 seen single eggs, attached like the others by their side to the supporting 

 surface, which showed a distinct knob or amorphous tail posteriorly. 

 I have no note now b}^ me of there being in such cases any reduction 

 in the amount of cement at the margin, though one would expect 

 this to follow. 



{h) With the eggs are found numerous empty shells from which 

 the operculum is generally pushed completely off. The operculum 

 frequently is seen persistent in recently hatched examples, so that its 

 removal in other cases may be due to the movem.ent of the insects 

 crowded into a narrow space. With the egg-shells are to be found 

 cast-off skins of all stages, and both are frequently devoured — pre- 

 sumably by the parasites themselves, though mites may also be 

 responsible for this. 



((?) There are also large accumulations of faecal matter in the 

 form of a fine grey sand which may till the proximal third or more of 

 the calamus. 



Putting these observations together one may hjq^othetically recon- 

 struct in outline the life-history of the parasite. Access to the quill is 

 gained by the Golpoceplialum boring into the rhachis, several specimens 

 using the same hole. Once inside the quill the parasites feed on the 

 pith and " seele," attain maturity and breed freely. Either through 

 failure of their food supply, or through overcrowding, an exodus is 

 initiated, the oi-iginal entrance being enlarg-ed or a new hole formed. It 

 is possible, of course, that before any general exodus takes place the 

 adult parasites may have left and returned to their strange habitat many 

 times. The very thorough examination of the plumage of several hosts 

 made, renders this highly improbable. I think it likelj^ also, that new 

 colonies are initiated, not by adult females, but by immature examples. 

 I have recorded these observations, incomplete as they are, partly 

 because of their — to the possibl}-^ prejudiced eye of a student of the 



