Botany. 333 



as follows : they conform closely to the usual formula, 3RO, R 3 

 3Si0 o and show that this garnet belongs, as was to have been 

 expected, to the almandite or iron-alumina, variety of the species* 





I. 



II. Mean. 



Molecular ratio. 







Si0. 2 



39-31 



39-27 39-29 



-r- 60 = 





•655 



3 



A1 2 3 



21-73 



21-67 21-70 



-4-102 = 





•212 



1 



Fe a 8 



tr. 



tr. tr. 











FeO 



30-67 



30-97 30-82 



-r- 72 = 



•426 ^1 







MgO 



5-22 



5-30 5-26 



-f- 40 = 



•130 ! 

 •035 f 



•613 





CaO 



1-95 



2-04 1-99 



-T- 56 = 





3 



MnO 



1-57 



1-46 1-51 



-f- 71 = 



•022 J 









100-45 



100-71 100-57 





Sp. grav. 



4-095 



4-091 











Sheffield 



Laboratory, 



March 6, 1891. 















III. Botany. 









1. The West American Oaks ; by Professor Edward L. 

 Greene, illustrated from drawings by Mr. George Hansen, 

 part II, San Francisco, 1890, quarto of 31 pages with 13 plates. — 

 The first part of this extended work, already reviewed in these 

 pages by Professor Goodale, appeared nearly two years ago, 

 describing and illustrating more than twenty oaks of the western 

 States, and defining several new species and varieties. -As a 

 prefatory note in part ii explains, Mr. James McDonald, through 

 whose liberality the work is being published, became impressed 

 with the desirability of further investigation of the new and 

 doubtful forms described in the first part. Accordingly he gen- 

 erously defrayed such traveling expenses as were necessary to 

 secure further material and information ; and Professor Greene 

 spent the summer of 1889 in studying the oaks in various parts 

 of the west and in collecting specimens for illustration. The 

 results of these later observations are embodied in part II and 

 comprise a number of additions and several significant corrections 

 to the material of the first part. Thus Quercus McDonaldii, 

 Greene var. elegantula Greene, and Q. Morehus Kellogg, are now 

 regarded as probable hybrids, while Q. dumosa Nutt. var. poly- 

 carpa Greene appears to have been founded upon an abnormal 

 state of Q. dumosa. While no quality in a scientific investi- 

 gator is more to be desired than perfect frankness in confessing 

 and correcting errors, it cannot but seem that where several 

 such corrections have to be made, so shortly after the appearance 

 of a work, that its publication was premature. Unfortunately 

 some of the descriptions in part ii seem no more likely to be per- 

 manent. Thus a new species is described solely from the foliage 

 of sterile shoots. How uncertain and inexpedient such species- 

 making is, will appear from the description itself, a part of which 

 we may quote : " The almost orbicular general outline of the leaf, 

 and its deep, crowded and even imbricated, doubly lobed margin 

 are very striking peculiarities. But these are the leaves of sterile 



