-ITS LIMITATIONS ee 229 
Cordaiteae, a type already very ‘perfect and: specialised.. We ‘cannot then 
draw from the data of Palaeobotany which we possess any indication of 
the origin of these - first Gymnosperms.” - This illustrates how hopeless it 
must be, so long as earlier strata yield only indefinite remains or none at 
all, to base upon stratigraphical evidence any consecutive story of the 
rise of a Land-Flora: for on a comparative basis these Gymnosperms which 
thus appear so early stand high in the: scale of Vascular Plants. Other 
examples might be quoted, but this will suffice to illustrate the deficiency 
of the record as regards prime origins. It has already been noted that 
developmental detail is usually absent from fossils, and that many are 
known only as impressions, without the possibility of minute structural 
examination of their tissues under the microscope. These considerations 
only show still further how scanty is the positive information from study 
of the fossils which is available for elucidating the early origin of the 
sporophyte. ; 
* There is also a- converse line of information, which involves negative 
evidence, based on the absence of certain types from strata where others 
are present. It may be held that organisms which first appear in the 
earlier strata are more primitive branches of the evolutionary tree than 
those which appear only in the more recent strata. But the fact that the 
record is, as we have seen, so very incomplete as regards the prime origins 
of the leading phyla will at once strike the note of caution in use of 
such negative arguments. Moreover, the probability of preservation of the 
representatives of any group may depend greatly on the character of the 
organisms in question: thus:it need be no surprise that the small and 
delicate Bryophytes are conspicuous by their absence from the earlier 
records, while Algae are but rarely preserved. Again, the non-representation 
ef any group may depend in some measure on the position in which the 
plants grew: thus the flora of uplands will be less likely to be preserved 
than that of low-lying lands or swamps; this argument has sometimes 
been applied in explanation of the absence of Angiosperms till a relatively 
late period. It is often possible to make out a plausible case from such 
negative evidence: but its insecurity is obvious. To use it with effect 
it must be supported by other considerations, such as argument from 
comparison. Thus the absence of evidence that Polypodiaceous Ferns 
existed in Palaeozoic times, must be taken with the position which is 
assigned to them on grounds of comparison among other Filicales; it 
then becomes a very convincing argument as showing their later derivative 
character, and the more so that Fern-types are among the best-preserved 
of early fossils. On the other hand, the entire absence of well-authenticated 
Ophioglossaceous remains from all the earlier formations only intensifies the 
difficulty of the problem which surrounds these curious plants, and cannot in 
itself be accepted as demonstrating that they are of relatively recent origin. 
These remarks-are intended to indicate the limitations to which the use 
of palaeophytological evidence must necessarily be subject. It is when these 
