250 Scientific Intelligence. 



infusible with phosphorus salt, reacts for titanic acid and iron. 



Two analyses of carefully selected material gave the following 



results : 



Ti0 2 90-78 9080 



FeO 8-10 7-92 



Sn0 2 1-32 1-38 



MnO trace trace 



100-20 100-10 



The iron is assumed to be present as FeO, but the state of oxi- 

 dation was not determined. The presence of Sn0 2 in the min- 

 eral was known before the material for these analyses was selected 

 and with the greatest care no cassiterite could be recognized in 

 the mass; as the larger number of the fragments showed portions 

 of crystal surfaces, and as the material is easily and completely 

 decomposed by potassic hydric sulphate it is very improbable 

 that the stannic oxide in the analysis is due to the presence of 

 admixed cassiterite. Since the crystals did not bear any obvious 

 resemblance to the known forms of titanic oxide, they were sent 

 to E. S. Dana and were identified by him as twins of rutile, un- 

 usual in form as they are remarkable for their high specific grav- 

 ity. Since then they have been more minutely studied by Mr. 

 Pirsson, who gives the following account of them. 



Note by L. V. Pirsson. — The crystals, of which a series of 

 seven was examined, are not suited for accurate measurement, the 

 faces giving faint and diffused reflections of the signal of the 

 goniometer and sometimes no reflection at all ; this is due to the 

 fact that the surfaces are all more or less etched and pitted, while 

 the polished points between are curved or broken. By using a 

 condenser and the d ocular of Websky enough light was collected 

 in several cases to give signals which afforded fairly accurate 

 measurements. These showed the following forms to be present : 

 a, i-i, (100); e, 1-i, (101); s, 1, (HI). These were identified by 

 the following calculated and measured angles : 



Calc. Meas. 



e^ a (101 * 100) 57° 12f 57° 10', 57° 20' 



a~ a (100 ~ 100) 65° 34|' 65° 50', 66° 08' 



5^s~(111-iTl) 56° 52-1' 56° 50', 56° 45', 56° 38' 



There were also a few other forms present on some of the crys- 

 tals, but these could not be identified with any certainty for the 

 reasons mentioned. The apparent orthorhombic symmetry of 

 these crystals, due to the extension of one pair of unit pyramids 

 and the absence of the other two is shown in the accompanying 

 figure. It illustrates the largest crystal examined, which is about 

 3mm b roac i an( j w ide by nearly 2 mm thick. That end of the crys- 

 tals where the re-entrant angle due to twinning should appear is 

 always broken and imperfect as if they had been attached at that 

 point. Somewhat similar forms of rutile have been described by 

 Miklucho-Maclay, Jahrb. Min., ii, 88, 1885. 



