174 Woodruffe-Peacock : Thrush Stones, etc. 



(Risso) unbanded. This is followed closely by three others in 

 order: — 112234425 to 112153425. Then come the specimens 

 more ' lightly ' banded on the upper side. The confluent type 

 {i 12234425) is a fairly common shell at anvils on fresh-water 

 alluvium, but the other form of it, where the bands show as 

 deep black on a dark brown ground is rare. It is apparently 

 the best protected form we have, for where it abounds it is 

 not frequently taken. Confluent mouthed shells are more fully 

 protected than plain banded ones. Soils and localities vary 

 greatly in banding and interspacing formulae ; and yet there 

 is a strong family likeness in shells from one spot as we should 

 naturally expect. So much is this the case that with sufficient 

 notes on local shells, and a well-arranged register, I believe it 

 would be quite possible to say from what soil, if the register were 

 kept under localities, I could almost say from what spot, a given 

 box, with a sufficient number of specimens, had been taken. 



The following notes may be found useful. One method of 

 reading off and recording the banding and interspacing must be 

 followed. Turn the shell bottom upwards, in the dextral 

 type with the mouth to the left hand. Draw an imaginary 

 line from the point where the lip joins the body whorl, through 

 the umbilicus round the shell, and read the banding and inter- 

 spacing off along this line. 



In every shell practically, unless some abnormality is found, 

 the third band is always the longest, or approaches nearest to 

 the lip. When the other four bands are absent, and the third 

 band is abnormally developed, there is generally a point extend- 

 ing beyond the average length of this wide band toward the 

 lip, approximately three twenty-fourths of the entire band 

 space wide, indicating the position of the original third type 

 band. Here is an instance. A ' dead ' specimen of rubella 

 (Moq.) 112162425, from road hedge side of pasture on sandy 

 glacial gravel. Three interspace band widths beyond the nor- 

 mal had been covered, one above and two below, as the longer 

 extending original third band indicated. On average shells, 

 on the soils I have worked, the bands come in order of length 

 towards the lip as follows :• — 34215. Exceptionally we find 

 32145 ; more rarely still some other order. The third band 

 too, usually turns slightly down at the lip. When the two 

 lower bands are confluent, there is generally a slight tendency 

 to bend upwards, just before finishing, though the lip ending 

 itself is generally in the normal line. 



[To be continued). Naturalist, 



1 



