May 5, 1922] 



SCIENCE 



483 



present concisely the essential steps in the 

 processes under consideration. These, with 

 other features, make the book very accessible 

 and helpful. It might here be suggested that 

 the diagrams would be improved by larger 

 index characters, and that somewhere a concise 

 index to the various terminologies scattered 

 through the chapters would make them more 

 available. 



It is not to be hoped that in a book of this 

 character there should be an absence of errors, 

 although in this instance they are not so 

 numerous as usual. Certainly they do not 

 render the text as a whole unsafe for the un- 

 guided beginner. Because of the merit of the 

 book in general and its obvious adaptability to 

 the present needs of a great variety of people, 

 it is particularly important to reduce errors of 

 all sorts to a minimum. Undoubtedly, the cor- 

 dial invitation of the author for assistance in 

 eliminating these will be met with a helpful 

 response bj' his fellow workers. Here it should 

 suffice to speak of only more general features 

 needing attention. 



Owing to the fact that the book will most 

 largely be used by those generally unfamiliar 

 with cytology, and having varied approaches 

 to it, there is need for the greatest clearness iu 

 distinguishing between the diii'erent categories 

 of objects and conditions described. This is 

 not- always done and there is sometimes con- 

 fusion between gene and character, and be- 

 tween the valence of the elements in the 

 chromosome complex. In the effort to simplifj' 

 the presentation of the maturation phenomena 

 in some of the diagrams, only one mitosis is 

 shown. While this displays clearly one of the 

 important conditions of meiosis it entirely 

 neglects another, viz., the essential unity of the 

 two matui'ation mitoses as a process. This is 

 further emphasized by the consistent use of the 

 tei'ms "heterotypic" and "homotypic." Enough 

 evidence has been presented to show beyond 

 question that the first maturation mitosis is not 

 necessarily a reduction division as the terms 

 imply. It is necessary only to recall the be- 

 havior of the sex chromosomes in the Hemip- 

 tera and the "selected chromosomes" in Phry- 

 notittix, as described by Wenrich, to demon- 

 strate this. There is something in meiosis be- 

 sides a reduction division and an ordinary 



equation division. It is important to show 

 clearly that meiosis is a unique phenomenon. 



Doubtless, there are other instances of sim- 

 ilar differences in point of view between 

 author and reviewer which might be used to 

 illustrate the present status of opinion in 

 cytology, and the degree of adaptability of the 

 test of Professor Sharp as an introduction to 

 the subject. What has been given will, how- 

 ever, suffice to show that the existing differences 

 of opinion are not extreme, that they are fairly 

 presented in the text, and that in their exposi- 

 tion, a work has been produced that will serve 

 to extend the usefulness and influence of 

 cytology greatly. It is not venturing far to 

 predict that the "Introduction to Cytology" 

 will take its place as a worthy member of the 

 very successful series of which it is a part. 

 C. E. McClung 



SPECIAL ARTICLES 



CONTINUOUS RENEWAL OF NUTRIENT 

 SOLUTION FOR PLANTS IN WATER- 

 CULTURES 

 In the experimental study of the salt nutri- 

 tion of plants, it is of course very important 

 that all the influential features of the culture 

 media be definitely known. The initial com- 

 position of a mixed salt solution employed for 

 water-cultures may be known with a marked 

 degree of accuracy, but the chemical make-up 

 of such a nutrient solution begins to be altered 

 immediately after the introduction of the 

 plants; materials, of course, move from the 

 roots into the solution, as well as in the oppo- 

 site direction, and the solution soon becomes 

 significantly different from what it originally 

 was. Since there is no feasible way by which 

 all the various kinds and rates of alteration 

 may be adequately determined, the culture 

 solutions must be renewed from time to time if 

 the growth of the plants is to be correlated 

 with known chemical conditions surrounding 

 their roots, and renewal must be frequent 

 enough to allow these unknown alterations to 

 be regarded as uninfluential. 



How frequently water-culture solutions 

 should be renewed is always a difficult ques- 

 tion. With small culture vessels, with large 

 plants, or with many plants in a vessel, it is 



