249 
Distribution of the Rhabdophora. 
they were present, though in far less abundance, in the 
“ Colonies ” imbedded in the schists of D d 5, the highest 
zone of his Inferior Division. The majority of the Bohemian 
forms were found to occur in the Coniston Mudstones ; and 
the natural inference was drawn by many palaeontologists that 
Barrande had placed the boundary between his two divisions 
at a- lower systematic horizon than that of Murchison—an 
opinion Murchison himself appears to have regarded with 
favour (‘ Siluria,’ p. 374). 
In North America the first prolific graptolitic formation (the 
so-called Hudson-River group), detected by the New-York 
geologists, was the highest natural group of rocks that could 
with propriety be assigned to Murchison’s Lower Silurian. 
It thus occupied a systematic position exactly equivalent to 
that of the Coniston Mudstone and the band D d 5 of Bo¬ 
hemia. The much grander series of graptolitic rocks (the 
Quebec group) afterwards discovered by the Canadian geolo- 
logists, were at first placed upon the same parallel; but the 
subsequent discovery of primordial genera within them led to 
their relegation to the base of the Nevv-York system, into the 
exact place of Murchison’s Llandeilo formation. 
Thus, on both sides of the Atlantic, it appeared evident that, 
as regarded their vertical range and specific development, the 
Graptolites presented two distinct maxima—the one near the 
base of the Ordovician system, the other near its summit. 
Their total range appeared to be coincident with the limits of 
Murchison’s original Silurian system; but, with the exception 
of a few scattered examples, they were wholly restricted to 
these two well-marked horizons. . 
The two maxima thus recognized in Britain and America 
were by no means, however, regarded as of equal importance. 
That at the summit of the Caradoc sank into comparative 
insignificance when compared with the maximum in the 
Llandeilo; and a tendency was soon developed among pale¬ 
ontologists to refer every prolific Graptolite-bearing stratum 
to the Llandeilo formation. Some of the examples of this 
tendency are very curious, as illustrative of the extent to which 
even the most cautious investigators will allow themselves to 
overlook or disregard facts when they stand in opposition to 
what appears to be a well-grounded generalization. 
In the paleontological portion of Ramsay’s ‘Geology of 
North Wales ’, Mr. Salter assigns the black shales of Conway 
to the Llandeilo formation for the sole reason that, like the 
Llandeilo of South Wales, they contain Graptolites in more 
than ordinary abundance — and this in the face of the admitted 
fact that they are surrounded on all sides by strata either of 
