174 Procfedingf of the Royal Irish Academy. 



observed at all in some (hedgiiigs wliere the long form occurs. At Tan 

 Buoy, Millport, the proportion of long to short is about 50 to 1. 



Measurement of the primordial chamljers reveals the striking fact that 

 the globular primordial chamber of the long dwarf form averages about 

 15^. It will be observed that this agrees with the average primonlial of 

 the microspliere of the nonual form, and we are therefore faced with the fact 

 that a tapering shell is connected with the megalosphere in the normal form, 

 but with the miciosphere in the dwarf. As with the normal form, there are 

 exceptional specimens in which the microsphere measured as little as 12/i 

 and as much as 18//. 



Tlie measurements of primordials of the short dwarf form offer no solution 

 to the problem. Of the relatively few specimens available, some agreed with 

 the long form, having a diameter of 15/i, but other specimens gave a 

 diameter of IS/i, 20/i, 25|i, the last being nearly double the average size, and 

 noticeably large and thin walled. As already stated, the primordial is 

 frequently missing in the short form, though very rarely wanting in the 

 long. What the significance of these differences may be we cannot at 

 present say. The only localities from which these pygmy forms have been 

 recorded are as follows : — 



(i) Greenland, 35-50 m. {passim, Goes, 1894). 

 (ii) Pacific Ocean, 995 fms. Goes, 1896. 

 (iii) Clare Island (W. of Ireland,\ H.-A. and E., 1913. 

 (iv) Millport (off the Tan Buoy), H.-A. and E. 

 |(v) Loch Striven, 70 fms. (W. Scotland), H.-A. and E. 

 (vi) Gulf of St. Lawrence (Canada), 212 fms., H.-A. and E. 

 (\-ii) Terra Nova Stn. 355 (Antarctic Benthos, 13), 300 fms. 

 H.-A. and E. 



We have never observed the dwarf in the living state or found it in our 

 tanks. This dwarf, or pygmy, form cannot be confused with young specimens 

 of the noi-mal form. The latter occur in quantity in any material in which 

 the species is common, and exhibit as a rule a primordial chamber, and one, 

 or perhaps two, triserial groups of chambers (fig. 14). They are, exteriorly, 

 rough in texture, and cannot be mistaken for the smooth and many-chambered 

 dwarf V. piisilla Goes. 



The marked tendency of V. polystropha to select and incorporate heavy 

 minerals among the nonual siliceous sand gi-ains, of which it constructs its 

 tests, presents a biological problem which is still far from solution, but the 

 fact stands out with striking prominence (figs. 15, 16). We have referred 

 above to our eailiest observations of this phenomenon, which aroused con- 



