434 



SCIENCE. 



[N. S. Vol. IV. No. 91. 



They are found on roots growing near tlie 

 surface and in great numbers, especially in 

 very damp soil. In development they differ 

 from ordinary lenticels, owing to absence 

 of initial stomata. Plants growing in sim- 

 ilar localities, such as Impatiens Julva, Myrica 

 gale, etc., never show such adaptation. 

 The question arises as to the significance of 

 these changes, which appear to be constant 

 in the cases mentioned, and experiments 

 on plants grown in water may serve toward 

 solving the question. 



A comparison of the flora of Erie Co., Ohio, 

 with that of Erie Co., N. Y. : By E. L. 



MOSELBY. 



Though both counties are adjacent, the 

 Ohio district, so far as now known, con- 

 tains 265 native species of phanerogams, 

 not known to occur within 50 miles of Buf- 

 falo. The causes are said to be mainly 

 climatic, the average date of the last killing 

 frost in spring at Sandusky being April 

 30th, at Buffalo May 20th; and the first 

 frost in autumn at Buffalo being September 

 15th, at Sandusky October 24th, One of the 

 causes of the later spring at Buffalo is the 

 blowing of the ice to the east end of the 

 lake. 



Sporojphyll-transformation in. dimorphic ferns : 



By Geo. F. Atkinson. 



The author detailed the results of experi- 

 :inents conducted on Onoclea sensibilis and 0. 

 :struthiopteris, showing that cutting off the 

 vegetative leaves in the middle of May and 

 :again in June would cause the partial or 

 complete transformation of the young sporo- 

 phylls to vegetative leaves. The paper 

 was illustrated by lantern views. 



The significance of the compound ovary: By 



Charles E. Bessey, 



In the study of the gynoecium of Angio- 

 sperms we are forced to conclude that its 

 primitive condition was apocarpous — in 

 other words, that whether monocarpic or 

 polycarpic there was at least no union of 



ovary with ovary. The original ovary was 

 doubtless simple. * 



By a comparative stud^^ of the ovaries of 

 existing plants we are led to the conclusion 

 that the syncarpous gynoecium was de- 

 rived from the apocarpous gynoecium. This 

 is so plain that it is needless to dwell upon 

 it. Both phylogenesis and ontogenesis fur- 

 nish us with numerous illustrations of the 

 truth of this statement. 



It is to be observed that the compound 

 ovary is a comparatively stable structure, 

 and that it changes slowly within any nat- 

 ural group, or in passing from group to 

 group. No part of the plant is more stable. 

 Yet with all its stability it undergoes 

 changes in certain directions. It is a com- 

 mon occurrence to find a pentacarpellary 

 ovary reduced to four, three, or two carpels, 

 and this may proceed until, as in some of 

 the Caryophyllacese, we seem to have but 

 one carpel remaining. In rare eases there 

 appears to be a reversion from syncarpy 

 towards apocarpy, as in the Apocynacese 

 and Asclepiadacese, but as a rule it may be 

 said that syncarpy once attained by a 

 group is persistently maintained, however 

 much of simplification it may otherwise 

 have undergone. 



The ultimate development of the com- 

 pound ovary is in the direction of a simpli- 

 fication of structure. Thus the many car- 

 pels of most Thalamiflorse and Heteromorse 

 are gradually reduced to the two carpels of 

 the Bicarpellatse. In the Calyciflorse the 

 Bosales and Myrtales have generally sev- 

 eral to many carpels, while in the Umbel- 

 lales there are but two. Likewise in In- 

 ferse the ovary in the lower group, Eubiales, 

 have more carpels than we find in the 

 highest group, Asterales. A similar sim- 

 plification occurs in the Monocotyledons, as 

 we pass from the Caronariese to the sedges 

 and grasses. 



This simplification of structure results in 

 increased parental care of offspring. Thus 



