50 KNUT LUNDMARK, GLOBULAR CLUSTERS AND SPIRAL NEBULJE. 



TT 



where V t = - F TO 32 , we obtain, as V m according to Table IV is 490 km 7 S ec, the following 

 three determinations of z m : 



0",o36 0".ooo356 (from Wirtz' 98 small nebulse) 

 0,033 319 (from Curtis' 66 spiral nebulse) 



0.029 0,oooi78 (from the proper motions of 7 spiral nebulse). 



Certainly these proper motions, and, consequently, also n m , are too great. 



Another way of determining n m is obtained by comparing the speed of the 

 solar motion F in relation to the system of spiral nebulse resulting from our Apex- 

 determinations with the value of the same quantity obtained by Wirtz from proper 

 motions of the 98 small (spiral) nebulse mentioned above. 



According to this we find: 



% m = 0", 000163. 



6. Measurements of Rotations and Internal Motions of the Spiral Nebulce. 



Several attempts were made during the 18th century to determine internal 

 motion in spiral nebulse but always gave negative results. With the development 

 of spectral analysis it became possible to detect the rotation of spiral nebulse a few 

 years ago. In 1914 thus Slipher 210 discovered rotations in the spindle nebala N. G. C. 

 4594, and at the same time Wolf 250 found in M 81 a radial velocity component due 

 to rotation. Shortly afterwards Pease 139 communicated the discovery of internal 

 motions in M 33. 



By measuring in the stereocomparator four pairs of plates taken at the Lick 

 and Mount Wilson Observatories during the interval 1899 — 1915, van Maanen I21 has 

 succeeded in establishing an annual rotational component of 0".022 in M 101 at a 

 mean distance of 300" from the centre, from wich follovvs a rotation period of 85000 

 years for a parti cl e situated at the said distance from the nucleus. 



Kostinsky 107 has communicated that he has by means of the same method 

 found internal motion in M 51 and that the motions amount on an average to 0",o4 

 — 0",os yearly. Lampland has also tried to derive internal motion in the same ne- 

 bula, and considers that he has found from his measures, which he, however, admits 

 to be very uncertain, that the secondary nucleus (N. G. C. 5195) of this nebula per- 

 forms a revolution about the nucleus in a period of 43000 — 108000 years. 



Supposing that the above components of rotation are on an average of the 

 same order of magnitude as the components of rotation found by measuring the 

 radial velocities, we obtain a means of estimating the average parallax of the nebulse. 

 The following results regarding the component of rotation in the direction of the 

 vision-radius have hitherto been published. 



