304 



C. ANDERSON. 



The irregularities in this list are not large and several 

 are in the wrong direction, if support for Goldschmidt's 

 view is sought. Of pairs which are doubtfully in twin 

 position we have the following: — 

 Segments. Meas. b \ b 



V, XII 62° 27' 



VII, VIII 62 32 



V, VIII 61 45 



Here the divergence from the theoretical angle is in the 

 right direction, but we cannot lay much stress on it. 



Group IX (Plate LIII, fig. 1). — The small group, \\ cm. 

 approximately in length, has a general resemblance to the 

 last described. Here I and V, and perhaps II, are para- 

 genic twins, the others metagenic. The forms represented 

 are b (010), m (110), r (130), x (012), k (011), i (021), v (031), 

 y (102), p (111). 



Segment. 



v 



I 



329° 9 



II 



32 1 



III 



143 38 



IV 



94 48 



V 



26 27 



VI 



86 42 



Limits. 



329 6 

 31 56 



143 37 

 94 43 

 26 24 

 86 41 



11 



7 



329 

 32 



143 40 

 94 53 

 26 31 

 86 42 



Number of 

 Observations. 



5 

 5 

 3 



10 



2 







62 52 



174 29 



125 



57 

 117 



39 

 18 

 33 



Twin relations: — 



Segments. Meas. b A b 



I, v 



57° 18' 



I, II 



62 52 



I, YI 



62 28 



II, IV 



62 47 



II, V 



62 49 



Gale, for r-twin 57° 18'. 



Gale, for m-twin 62° 46' 



The segment VI is the only one which shows a notice- 

 able departure from the true twin position; unfortunately 

 only scanty measurements were obtained but the signals 

 were good and 18' is too great a discrepancy to be attributed 

 to observational error. Crystal VI, if twinned at all, is 



