916 



SCIENCE 



[N. S. Vol. XXXI. No. 806 



Some Aids to Better Work in Soie>ice: C. W. 

 Edwards, Trinity College, Durham, N. C. 

 (Read by title.) 

 A new Hybrid Habenaria of North Carolina-: 

 J. G. Hall, North Carolina Agricultural Ex- 

 periment Station, West Raleigh, N. C. 

 A hybrid Habenaria was reported from the 

 neighborhood of Kinston, N. C. This natural 

 hybrid seemed to be pretty well intermediate 

 between the two supposed parents H. ciliaris and 

 H. blephariglottis. Photographs of the flowers 

 were shown and these presented some characters 

 of the parents and the hybrid. 

 The Present Status of the Darwinian Hypothesis: 

 W. L. PoTEAT, Wake Forest College, Wake 

 Forest, N. C. 

 Some Experiments on Ionization by Impact: The 

 Time Variation of a Current through a Gas 

 Ionized by Radium: J. Blanchakd, Trinity 

 College, Durham, N. C. 



The ionization vessel was a glass tube with 

 parallel plate electrodes about five centimeters in 

 diameter, both plates coated (though unequally) 

 with a thin layer of a very impure salt of radium. 

 With the plat«s about one centimeter apart, and 

 the pressure about one millimeter, with a poten- 

 tial difference sufficient to produce considerable 

 ionization by impact, it was found that the cur- 

 rent decreased with the time the battery key 

 remained closed, reaching its minimum value in 

 about an hour. On opening the key the initial 

 conductivity was almost totally regained in about 

 the same time. Upon reversing the potential at 

 the end of an hour the current was sometimes 

 found to be greater than it was initially in this 

 reverse direction, but also decreasing with the 

 time as before. 



The potential difference apparently causes an 

 increased amount of ionization near the positive 

 plate. 



Further experiments are in progress. 



Is the Pusarium which Causes Cowpea Wilt 

 Genetically connected tmth Neocosmospora? B. 

 B. HiGGiNS, North Carolina Agricultural Ex- 

 periment Station, West Raleigh, N. C. 

 In 1889 the wilt disease of cotton was studied 

 by Professor Geo. F. Atkinson and its causal 

 fungus named Pusarium vasinfectum. A few 

 years later (1894-99) the wilt disease of cotton, 

 watermelon and cowpea was studied by Erwin F. 

 Smith. He found no specific differences between 

 the fungi upon any of the three hosts. He found, 



however, upon some of the plants previously 

 killed by the wilt fungus, an aeigerous fungus 

 which he considered the perfect stage of Pusarium 

 vasinfectum. The fungus was therefore renamed 

 by him Neocosmospora vasinfecta, and this con- 

 clusion has been accepted by subsequent writers. 

 The evidence upon which this conclusion was 

 based was very weak, however; and a recent study 

 oi the two forms by the writer — the results of 

 which will at an early date be published in bul- 

 letin form — has caused the writer to reopen this 

 question which was considered closed. 



Some Experiments in the Propagation of the 

 Diamond-bach Terrapin: Heney D. At.t.kb, 

 Fisheries Laboratory, Beaufort, N. C. (Read 

 by the secretary.) 



This paper appears in full in the current num- 

 ber of the Journal of the Elisha Mitchell Scien- 

 tific Society. 



The Present Status of th^ Relativity Problem: 

 C. W. Edwards, Trinity College, Durham, N. C. 

 (Read by title.) 



The Locus of a Moving Point when the Sum of 

 its Distances from Two Fixed Points, their 

 Difference, their Product or their Quotient is 

 Constant: John F. Lanneau. 

 The loci determined by the first three conditions 

 are the well-known ellipse, hyberbola and lem- 

 niscate. 



Under the fourth condition: Take line through 

 the fixed points F and F' as a;-axis; the point 0, 

 midway between them, as origin; 2o for distance 

 F to F' ; K for the constant quotient when the 

 moving point is on one side of the i/-axis, and 

 therefore \/K the quotient when it has the cor- 

 responding position on the other side. 

 1. The equation of the locus is 



' -t- y= =F 2c 



K^ + l 

 K^ — 1 



I- -I- c« = 0. 



The locus, therefore, consists of two equal circles 

 whose centers are on the OJ-axis beyond F and F', 

 at equal distances from 0. 



2. A discussion of the equation shows: 



If £: = 1, the circles are of infinite radius, and 

 are tangent at 0. 



If K is either or oo, the circles reduce to the 

 points F and F'. 



If K has, in turn, any series of values between 

 1 and 0, or between 1 and co, the loci form a 

 group of circles about P and a similar group 

 about F'— the number of circles in each group 

 limited only by the number of values given to K. 



