J. Starkie Gardner — Development of Dicotyledons. 159 



Each new discovery should fall into a definite place, fill in some 

 gap, and serve as a new point of dejDarture. 



That this is not the case when new fossil floras are described is 

 but too apparent. When we sum up and examine the whole of the 

 knowledge we possess of the ancestral forms of existing Dicotyledons, 

 we find we are in a state of almost complete ignorance regarding 

 them. We merely know that in certain formations, which we believe, 

 from their included marine fauna, to be of Cretaceous age, perfectly 

 developed leaves of Dicotyledons abound ; but we do not know with 

 any certainty how far we are justified in assigning them to existing 

 genera, nor in what way they are linked to these. We know abso- 

 lutely nothing of the successive steps by which even the leading 

 existing families have been developed. We know nothing concerning 

 the evolution of the Leguminosge, the Proteacege, the Cupuliferae. 

 We see nothing beyond a more or less vague recognition of the fact 

 that plants with simpler flowers should have preceded those with 

 flowers of more compound character. We assume from the absence 

 of Dicotyledons in older Cretaceous rocks that they must have been 

 relatively somewhat rapidly developed ; but the where, how, and 

 whence, are still sealed books. 



In the first place, we must inquire whether we can be absolutely 

 certain that the ages of the Cretaceous beds which contain them are 

 correctly known. Our first impression would be that the age of the 

 Aix-la-Chapelle series is established on the most unequivocal evidence, 

 for the beds with plants are intercalated between strata which belong 

 lithologically and at first sight paljeontologically to our Upper Green- 

 sand and Chalk, yet nevertheless the similarity may be completely 

 misleading ; and when we compare the faunas together critically, we 

 shall realize that those of the Aachenian series are relatively con- 

 siderably the younger. In an area of prolonged and gradual 

 subsidence, as Europe undoubtedl}' was throughout the XJjoper 

 Cretaceous period, the deep sea, or ancient gulf of the Atlantic, 

 as represented by the Chalk, would gradually encroach further and 

 further in every direction as the levels were lowered. At certain 

 depths blue mud resembl ing that known as the Gault, and green mud and 

 sands, similar to our Upper Greensand, would now be forming. These 

 require a less depth and are more littoral than the deeper sea deposits 

 corresponding with our Chalk, and in every area where chalk exists 

 are seen to underlie it. During the Cretaceous subsidence, where- 

 ever oceanic waters penetrated, one or other or both of these qualities 

 of mud must have preceded the deposition of true Chalk. Nothing 

 could be more erroneous and more liable to beget confusion than to 

 speak of all the blue mud as of one age, the green mud and green 

 sand of another, the calcareous mud with chloritic grains as another, 

 and the finer calcareous ooze as another. Their relative ages must 

 be judged by the faunas they contain, and the greater or less resem- 

 blance of the organisms composing these to the existing types now 

 found in the same qualities of sediment. There is no reason to 

 suppose that the Cretaceous subsidence extended over Europe syn- 

 chronously ; it is at all events perfectly obvious that so long as Chalk 



