138 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1025 



stinctively we turn to those objects to which we ascribe the greatest 

 distances from us. We hope that in the light coming to us from 

 these distant objects we may find something serving as testimony 

 of the long path it has traversed through this space. Speaking 

 metaphorically, from the dust of the journey we would infer some- 

 thing of the nature and direction of the path as well as the region 

 tln-ough which the journey lay. 



2. Which, then, are the most distant bodies? Through a process of 

 elimination we are quickly left with only two possible kinds of bodies 

 which mostly or wholly lie without that part of space wdiich is filled 

 with stars in varying number. Our sun is somewhat near the 

 center of this universe of stars. The boundary of it we may prac- 

 tically place where the star density is one one-hundreth that close 

 to our sun. To realize the vanishingly small density in which the 

 matter, collected in the form of stars exists, we may think of them 

 as grains of sand so scattered that each grain is distant about 100 

 kilometers (62 miles) from its nearest neighbor. This shows to 

 scale the relative emptiness of space. The limit of our universe of 

 stars as above defined lies so far away (30,000 light-years) that it 

 takes 30,000 years for light to come from the boundary to us. Be- 

 yond, outside the bounds of our stellar system (galaxy) lie numerous 

 globular star clusters and spiral nebulae — not all the clusters but all 

 of the nebulae. 



How may we obtain an idea or measure of these immensely great 

 distances? Direct trigonometric triangulation naturally can give 

 us no information since we have to do with annual parallaxes giving 

 angles as small as 0.000005 angular seconds, and indeed less. The 

 motions of the spiral nebulae give us the answer and a riddle at the 

 same time. For 15 years we have known that the lines in the spectra 

 of the spiral nebulae show shifts which have been interpreted as Dop- 

 pler displacements, a physical interpretation which is taken to indi- 

 cate and measure radial velocities of the nebulae. These velocities are 

 much greater than those generally found for the fixed stars. They 

 may be of the order of 2,000 kilometers (1,250 miles) per second. 

 At present radial velocities have been determined for some 50 spiral 

 nebulae. These have already been analyzed to discover interesting 

 systematic characteristics in the nebular movements. 



All apparent motions of the heavenly bodies are made up of two 

 components : Their own or " peculiar motion " and a mirroring in 

 them of the motion of the observer.- The latter varies for different 

 directions relative to our movement and with the distance of the 



2 Note by the translator. — An analogous motion mfay be observed from the window 

 of a rapidly moving train. The trees of the middle distance seem to recede, while the 

 more distant ones move along with the train. 



