550 



SCIENCE. 



[N. S. Vol. I. No. 20. 



upper part of the loioer Silurian; but on page 

 566 he says that the oldest known insect is 

 the Pakeoblattina of the itpper Silurian. 



We might mention others, but they are 

 all trifling. In fact, the accuracy of the 

 book is extraordinary. 



In conclusion, we must heartily and most 

 gratefully welcome the new edition. It is 

 hard to say what American would be with- 

 out Dana's Manual. Its encyclopedic full- 

 ness and yet extreme conciseness makes it 

 hard reading for those who come to it with- 

 out serious purpose. The word 'Manual ' ex- 

 actly expresses its purposes and uses. It 

 must be in the hands of every special stu- 

 dent; it must lie on the table of every 

 teacher of Geology to be consulted on every 

 subject of doubt. 



I had just finished this notice when the 

 sad news of Dana's death was flashed across 

 the continent. All recognized that this 

 event could not be long delayed; but none 

 the less it came as a shock to every man of 

 science in the country. "We are thankful 

 that he lived to finish this new edition, for 

 it is indeed the only fitting monument. No 

 monument is worthy of a man of science 

 except that which he erects for himself. 

 Joseph Lb Conte. 



University of Califoenia. 



A Sandhook of Systematic Botany. By De. E. 

 Warming. Translated and edited by M. 

 C. PoTTEE. 8vo. pp. 620, fig. 610. Lon- 

 don, Swan, Sonnenschein & Co. New 

 York, Macmillan & Co. 1895. 

 This excellent English translation of Pro- 

 fessor Warming's important work will be 

 welcomed by all students and it cannot fail 

 to have a wide use as a text-book. The de- 

 scriptions of the groups are clear, concise 

 and complete, the illustrations capital and 

 many of them original, and the press-work 

 leaves nothing to be desired. 



The arrangement of groups is from sim- 

 ple to complex — the only arrangement com- 



patible with our present knowledge. The 

 special application of this principle maj^ be 

 best stated in Dr. Warming's own words as 

 printed in the preface : 



" Each form which, on comparative morphological 

 considerations, is clearly less simple, or can be shown 

 to have arisen hy reduction or through abortion of 

 another type having the same fundamental' structure, 

 or in which a further differentiation and di\'ision of 

 labor is found, will be regarded as younger, and as 

 far as possible, and so far as other considerations will 

 admit, will be reviewed later than the 'simpler,' 

 more complete or richer forms. For instance, to serve 

 as an illustration : Epigyny and Peeigyny are less 

 simple than Hypogyny ; the Epigynous Si/mpetalse, 

 Choripeialie, Monocoiyleilones are, therefore, treated 

 last ; the Hydrocharitacex are considered last under the 

 HeloMem, etc. Zygomoephy is younger than AcTiNO- 

 MOEPHY ; the Sdtamineie and Oynandrss, therefore, 

 follow after the Lilliflorse, the Scroplmlariacex after 

 the Solanacex, Linaria after Verbascum, etc. FOEMS 

 WITH UNITED LEAVES indicate yoiinger tj-pes than 

 those with free leaves ; hence the Sympetalse come 

 after the Clwripetala:, the Silenex after the Alsineie, 

 the 3Ialvacex after the Sterculiacex and TiUacex, etc. 



"Acyclic (spiral-leaved) flowers are older than 

 cyclic ( verticillate-leaved ) vnih a definite number, 

 comparing, of course, only those ■n-ith the same funda- 

 mental structure. The Veronica-Typt must be con- 

 sidered as younger, for example, than Digitalis and 

 Antirrhinum; these again as younger than Scroplmlaria; 

 Verbasmm, on the contrary, is the least reduced, and, 

 therefore, considered as the oldest form. Similarly 

 the one-seeded, nut-fruited Banunctilacex are con- 

 sidered as a later type (with evident abortion) than 

 the many-seeded, follicular forms of the order ; the 

 Paronychiese and Chenopodiacex as reduced forms of 

 the Alsinese type ; and the occuiTence of few seeds in 

 an ovary as generally arising through reduction of the 

 many-seeded forms. The Cyperacex are regarded as- 

 a form derived from the Jiincacex through reduction, 

 and associated with this, as is so often the case, there 

 is a complication of the inflorescence ; the Dipsacacete 

 are again regarded as a form proceeding from the 

 Valerianacex by a similar reduction, and those in 

 their turn as an off-shoot froni the Caprifoliacew, etc. 

 Of course these principles of sj'stematic an'angement 

 could only be applied very generally ; for teaching 

 purposes they have often required modification." 



While there is wide difference of opinion 

 among botanists as to the relative degree of 

 complexity of some of the families, and the 

 sequence adopted by Engler and Prantl in 



