to the seed of B. Delavayi, as shown in drawings at Kew and in the "Pflanzenreich". 

 If the bract and seed prove to belong to different species, the bract is the type of 

 B. digitata. 



BETULA Sp. 



PI. IV, fig. 40. 



A female catkin, incomplete at either end, represents our common European 

 birches; but is too imperfectly preserved for determination. The catkin is small, and 

 appears to have been much shorter and more ovate than in B. pubescens or B. verrucosa, 

 though the oval outline may perhaps be due to compression; the bracts are small, thin, 

 strongly striate, and seem to have a blunt middle lobe. The reflexion of the bract, seen 

 in the photograph, may be due to pressure, the shape of the lateral lobes cannot be 

 made out, nor can the seed be seen. 



Length 5 mm., breadth 3 mm. Reuver. 



We do not recognise this catkin as belonging to any European birch. 



BETULA Spp. 

 PI. IV, figs. 41-43. 



Seeds of two other species otBetula (in addition to the seed provisionally referred 

 to B. digitata) have been found. From Reuver, where the catkin just described was 

 found, we have a large imperfect seed (not figured); but this seed seems too large to be* 

 long to the catkin, for it is nearly as wide as a whole scale. From Swalmen we have the 

 small obovate seed shown in fig. 41, which in size might belong to the catkin; but they 

 were found at different localities. 



Two male catkins found at Reuver (figs 42, 43) evidently belong to two quite 

 different species ; but we have no means of connecting them with the plants yielding 

 either the female cone or the isolated seed. 



ALNUS. 



Cones, bracts, and seeds of Alnus are abundant, and seem to fall into four groups 

 We have examined the cones of all the Palaearctic alders in the Kew Herbarium, but 

 none of these agree with any of our fossils, though A. elliptica Req. seems near to our 

 Species 1. We find the scales and seeds of living forms so variable, and so many recent 

 and fossil alders have been described, that we think it safest not to name the Reuverian 

 cones, but merely to describe them as nearly as we are able under species Nos. 1 to 4. 



ALNUS Sp. 1. 



PI. V, fig. 1. 



Cone oblong, length 12 mm., breadth 8 mm.; scales few large flat, front and 

 side lobes wing*like, truncate above, margins crenulate, scarcely sculptured; front lobes 



76 



