THE ORIGIN OF ANGIOSPERMS. 71 
These conclusions are easily esplained on the supposition that the earlier 
Angiosperms still retained the Cycad-like type of foliage of their ancestors ; 
and as our knowledge of the Mesozoic floras is in great part, if not almost 
entirely, derived from detached leaf-impressions, and not from fructifications, 
it is not surprising that we have been puzzled by the facts as presented 
by the geological record. The so-called “ sudden appearance” of the 
Angiosperms in Neocomian times may have no significance as regards the 
phylogeny of the group, but may well express the fact, that this group, 
already highly evolved and diversified, then assumed the free-branching 
habit and consequent mierophylly. This hypothesis also explains why this 
race appears to be dominant over other groups even in the Neocomian period, 
for the subsidiary incoming stage of the life-line would be masked by the 
retention of the Cycad-type of foliage. 
But, apart from these considerations, the great problem remains as to how 
the microphyllous foliage of the Angiosperms was derived from the Cycadean 
type. On this point we are at present unable to offer any suggestion unless 
we call mutation to our aid (see p. 36). So far we are not aware that fossil 
botany has afforded evidence of transitions from the type of foliage peculiar 
to any Mesozoic group to that of the Angiosperms. 
THE ORIGIN OF MONOCOTYLEDONS. 
It is still a matter of keen debate whether the Dicotyledons or the 
Monocotyledons are geologically the older group. The arguments are derived, 
partly from our knowledge of their living members, and partly from a study 
of fossil impressions. But it may be doubted whether either of these lines of 
attack afford sufficient data at present to settle the question quite beyond 
doubt. 
Some* have regarded the Dicotyledons as derived from the Monocotyledons, 
while othersf, including Hallierf, hold the converse, in some cases with the 
reservation that the Monocotyledons branched off from the main Angiospermous 
line, i. e. Dicotyledons, at a very early period. With the latter view we 
entirely agree. 
So far аз the fossil evidence is concerned, we doubt if it is possible to show 
that either group is really more aneient than the other. We agree with the 
opinion, now generally held, that the earliest fossil remains, whieh in the 
present state of our knowledge we can recognise as clearly belonging to the 
* Lyon (1901, 1905). 
t Sargant (1903), Mottier (1905), Chrysler (1906), Plowman (1906). 
і Hallier (1905). 
