October 23, 1914] 



SCIENCE 



G09 



far south as Alliance, whicli seems very prob- 

 able, the Ohio might easily have first broken 

 over here and have flowed westward.^ The 

 deep drift in this section makes it difficult to 

 determine this point but the width of the 

 Tuscarawas Valley, the narrowness of the 

 present Ohio VaUey and the occurrence of 

 lacustrine deposits north of Alliance and 

 mixed more or less with the drift in many 

 parts of the Grand River Valley seem to 

 strongly favor such an hypothesis. 



It is generally believed by geologists that 

 the preglacial divide of the Ohio drainage 

 was near New Martinsville, West Virginia. 

 A study of the direction of the streams along 

 the Tuscarawas would seem to indicate that 

 the preglacial divide along this stream was 

 near Port Washington and that Big Still- 

 water, Conotton and Big Sandy Creeks flowed 

 northwest, the former by Canal Dover along 

 the present course of Sugar Creek reversed at 

 least beyond Beach City. Whether this stream 

 joined the other two near Justus or Navarre or 

 flowed on northwest separately can not be 

 stated definitely because of the drift and the 

 changes brought about by the advance of the 

 ice. Below Port Washington the drainage 

 was probably westward into the Scioto Valley 

 and old Kanawha system. 



Upon the first advance of the ice southward 

 of Lake Erie the drainage of all northward 

 flowing streams was obstructed and it became 

 necessary for their waters to seek other out- 

 lets. As the country to the west was in gen- 

 eral lower the streams were dammed up until 

 they finally ran over the lowest divide on the 

 west. There was a tendency for them to follow 

 in a general way the ice border, just as the 

 Ohio and Missouri Rivers to-day follow rather 

 closely around the southern extension of the 

 ice. 



In view of the above considerations it is 



3 Since this article was written the writer has 

 had an opportunity to make further observations in 

 the country north of Alliance and has found fur- 

 ther evidence, particularly an old valley near 

 Eavenna, to substantiate the hypothesis that the 

 Ohio Eiver first broke over in this section and 

 formed the Tuscarawas-Scioto Valley. 



believed that the Tuscarawas-Scioto Valley 

 had its origin in an early advance of the ice 

 and represents the principal drainage line 

 during interglacial time, and that the advance 

 of the ice farther southward during the later 

 glaciations forced the Hocking, Muskingum 

 and possibly the Upper Ohio to change to 

 their present channels. Such a hypothesis 

 makes it possible to explain many very pecu- 

 liar connections between old valleys, which are 

 very difficult to understand otherwise. If the 

 time which elapsed between the different ad- 

 vances of the ice, had been estimated with any 

 d^ree of approximation it can be easily 

 understood how much larger valleys may have 

 been formed during interglacial periods than 

 since. The matter appears to deserve more 

 consideration in the interpretation of changes 

 of drainage than it has been given heretofore. 

 George N. CorrEY 



THE POISONOUS NATURE OF THE STINGING HAIRS 

 OF JATKOPHA UEENS 



Jatropha urens is one of the most abundant 

 Euphorbiaceous plants growing in or around 

 the savannas of the Pacific coast of Central 

 America. Its spread is favored by the fact 

 that the cattle avoid it, and because it is not 

 kept down by the too indolent owners of the 

 pastures. Everywhere it has the reputation of 

 being extremely dangerous, on account of its 

 poisonous effects. 



The plant is easily recognized: It is her- 

 baceous, 0.5 to 1.5 meter high, regularly rami- 

 fied, with large palmatilobate leaves, white 

 flowers and small, 3-eelled capsides. All parts, 

 trunk, leaves, flowers and fruits are covered 

 with long, hard and glossy, stinging hairs, 

 which protect the plant as barbed wire pro- 

 tects the fortifications of to-day. It would 

 seem as if the remarkable glossiness of the 

 stinging hairs might warn the curious against 

 approaching or touching. As a mater of fact, 

 the animals either by instinct, or on account 

 of the wisdom acquired through some previ- 

 ous experience, avoid contact with it. 



The vernacular name of Jatropha urens is 

 " ortiga " or " ortiga brava " (nettle) in Pan- 

 ama, and other parts of Central America, in- 



