May 21, 1897.] 



SCIENCE. 



789 



proof in any ease that they flowed out at 

 the surface as lavas. Like the gabbros, 

 they were developed in special centers and 

 preferentially in the areas already chosen 

 by the gabbros; hence, the black, rugged 

 mountains of gabbro are very generally 

 accompanied by more or less regularly 

 formed cones of granite and granophyre. 

 These two cones not only form a striking 

 contrast to the gabbros as regards their 

 contour, but also in their colors. They 

 range through various tints of yellow and 

 russet. The characteristic scenery of central 

 Skye, with the dark, rugged outlines of the 

 gabbros on the west and the smooth, pale 

 cones of the Red Hills on the east, depends 

 upon this difference of geologic structure. 

 The latest phase in the volcanic history 

 is marked by the uprise of another great 

 series of basic dikes which, like those of the 

 earlier time, follow a generally north- 

 westerly direction and rise even through 

 the latest masses of granophyre. On the 

 flanks of the Red Hills of Skye long, dark 

 ribs of rock may occasionally be seen even 

 from a distance, which mark the last efibrts 

 of the Tertiary volcanoes of Britain. 



W. F. MOESELL. 



SOME PEASES OF WEED EVOLUTION* 

 The common statement that " a weed is a 

 plant out of place "is by no means satis- 

 factory or final to the student. He is still 

 left to ask how it is that certain species 

 have such pronounced ability for getting 

 out of place. Almost any plant may acci- 

 dently get where it is not wanted, but com- 

 paratively few usually and persistently get 

 in the way. A bad weed species is aggres- 

 sive and persistent. "What qualities make 

 it so? 



As a partial answer I would suggest 

 four life attributes as follows : 1. Great 

 reproductive power. 2. Good provision for 



* Notes from a lecture before the University of 

 Termont Botanical Club. 



dissemination. 3. Various protective ex- 

 pedients. 4. Variability. 



The statement that a species, to rank as 

 a bad weed, should have large reproductive 

 power needs no elucidation. It is not even 

 necessary to demonstrate that weed species 

 are thus equipped. Dozens of striking 

 examples will occur at first thought.* We 

 should also remember that many of the 

 worst weeds have a double resource for 

 their multiplication ; i. e., they propagate 

 themselves by buds as well as by abundant 

 seeds. The Hawkweed throws out stolons ; 

 and a Canada Thistle chopped into 20 pieces 

 by the hoe becomes 20 Canada Thistles. 



It is also axiomatic that facility of dis- 

 semination is one of the qualities of a weed 

 species. 



Some weeds make headway by means of 

 special contrivances for avoiding disaster. 

 Certain ones can withstand hard drought. 

 Others have such deep roots as to be beyond 

 the reach of ordinary cultivation. The Dan- 

 delion, often cut down by the lawn mower, 

 bears its feathery tuft of seeds on a stem so 

 short as to escape the gardener. 



But variability is the ■ chief and most 

 significant quality of a weed species. 

 Moreover, variability is ultimate. And 

 those characteristics already mentioned are 

 all involved in variability ; for each one — 

 reproduction, dissemination, protection- 

 may vary, and through selection be indefi- 

 nitely modified. In this sense, however, 

 variability is not coordinate with the other 

 qualities mentioned. 



Other things being equal, then, that 

 species which is most variable is apt to be- 

 come a weed. This point may be illus- 

 trated, though perhaps not demonstrated by 

 reference to some statistics of weed floras. 

 Thus, starting with the proposition that 

 variability is roughly proportional to the 

 number of members in a systematic group, 



*cf. Kerner, ' The Natural History of Plants,' II., 



878. 



