LOWER SASKATCHEWAN RIVER VALLEY. 
15 
Conchidium decussatum Hall and Clarke, Pal. New York, 
VIII, Pt, II, 1893, p, 235, pi. 65, figs. 1, 2; pi. 66, fig. 15. 
Conchidium decussatum Whiteaves, Geol. Surv. of Can. 
Palseozoic Fossils, VoL III, Pt. IV, 1906, pp. 293-295, 
pi. 26, figs, 1-2. 
Conchidium decussatum Kindle, Geol. Surv. of Can., 
Summ. Rept. 1912 (1914) p. 249. 
The apparent limitation of the range of P, decussatum to 
the basal portion of the section at the Grand rapids of the Saskat- 
chewan, gives the species an especial interest from the strati- 
graphic standpoint. This, together with the extraordinary 
abundance of the shell in this horizon in the Saskatchewan 
River section, invites some further consideration of the variable 
characters of the species. In the numerous loose boulders of 
limestone in the old cut-off channel at Grand rapids and else- 
where in the vicinity this species is crowded together in such 
numbers that their separation except as fragmentary specimens 
is difficult. Other species are apt to be almost if not entirely 
excluded by this prolific shell from the beds containing it. 
Dr. R. C. Wallace has recently found this shell at Hilbre 
and Fisher Branch, Manitoba. The collection sent the writer 
by Dr. Wallace appears to indicate its abundance at these loc- 
alities which are more than 100 miles south of the Grand Rapids 
locality and near the base of the Silurian section. 
The figures of this species, previously published, show what 
their authors consider typical specimens, but give little concep- 
tion of the variability within its limits. Those which are 
here shown are intended primarily to illustrate the variable 
features of the shell. 
Conchidium decussatum belongs to a group of shells in which 
the specific characters are very plastic. The rather large number 
of specimens of this species secured by the writer, show that 
plasticity or variability is as characteristic of this as it is of some 
of the other pentameroid shells. Most of the observations which 
have been made on the variability of this shell relate to the ven- 
tral valve because the collection contains very few dorsal valves, 
the great bulk of the specimens being ventral valves. Hence 
