221 



The second volume with a portrait of Hermann Miiller, upon 

 whose studies the series is based, has now appeared. This volume 

 is special in nature and contains " an account of all known obser- 

 vations upon the pollination of the flowers of plants of Arctic and 

 Temperate zones." From the contents, however, we infer that 

 it is devoted to certain groups of dicotyledones ranging from the 

 Ranunculaceae to the Stylidieae. The sequence of the groups 

 is somewhat different from the arrangement with which we are 

 familiar and to give an idea of their arrangement and of the scope 

 of the work mention may be made of the more important orders 

 and families, employing the classification in common use in this 

 country : Ranales, Papaverales, Violaceae, Polygalaceae, Cary- 

 ophyllaceae, Portulacaceae, Hypericaceae, Malvales, Geraniales, 

 Sapindales and Rhamnales, Resales, Myrtales, Passifloraceae, 

 Cucurbitaceae, Cactaceae, Umbellales, Rubiales, Valerianales, and 

 Compositae which includes our Carduaceae and Cichoriaceae. 



The remaining families and orders will be discussed in the 

 third volume, while the fourth and final volume has been adver- 

 tised to deal similarly with plants outside Europe. The orders 

 mentioned above are not recognized in the discussion, the species 

 being arranged in fifty-six orders that correspond to our families. 

 For example the Ranales are discussed under six orders and 

 the Sapindales and Rhamnales are represented by six orders 

 promiscuously intermingled, fhe families (orders) of an alli- 

 ance are also very generally taken up in the reverse order of 

 their complexity. The rose group begins with the Leguminosae 

 and follows with the Rosaceae, Saxifrageae, and Crassulaceae, 

 though the Myrtales may possibly be cited as a reversal of this 

 treatment in which the reduced Halorrhagidaceae (Halorageae 

 in the text) are followed by the Melastomaceae, Lythrariae, and 

 Onagrariae. Just what purpose the authors had in mind by this 

 reversion of sequence — of going up the biological scale back- 

 ward so to speak — is not clear. We take it that an attempt has 

 been made to represent in this arrangement of the families, their 

 natural relationship. In some cases the sequence is a very prob- 

 able one, as from the Umbellales to the Rubiales, Valerianales, 

 and Compositae ; but in other cases the arrangement is unusual to 



