ADDRESSES 58 
by Dr. Moulton in tracing out the courses of the projectiles 
in the first nine test cases which he followed mathematically 
to see the actual workings of this combination of agencies. 
In these cases he made no postulates relative to the mass, 
position, orbit or velocity of the passing star, or of the pro- 
jectiles from the sun, that did not seem to lie within the pro- 
babilities of the case. In forty-eight cases later tried, he found 
the dispersing potency of the combination surprisingly effec- 
tive, even when the passing star was not more massive than 
our sun, and its approach not usually as close as that of the 
earth to the sun, though that distance and half that distance 
were used in a few cases. The remarkable fact was revealed 
that gravitation, commonly supposed to be the supreme agency 
of celestial concentration, is really, under the conditions o! 
swiftly approaching bodies of explosive habits, a very effective 
agency of dispersion. 
It was further shown that the passing star may remain 
within an effective range of action for a period of twenty 
years in certain cases. The average period in the cases se- 
lected proved to be about five years. 
During the effective period of influence a succession of 
explosive impulses must almost certainly take place, and these 
should give a stream of projectiles in the form of bunches or 
pulses of solar matter attended by diffusely scattered matter, 
the two elements constituting, by interpretation, the knots and 
haze so common in spiral nebulze. These streams would of 
course issue on opposite sides of the sun and would curve 
in opposite directions, a distinctive feature also common in 
spiral nebulez. 
It is not difficult to see that these features would arise 
naturally, if not inevitably, under these conditions, but it taxes 
the imagination somewhat more severely to follow the com- 
posite action through the whole series of pulsations called 
forth successively by the disturbing star until it passes to an 
ineffective distance, but if done, it will appear that the ar- 
rangement of pulses at any instant takes the form of a spiral 
with two streams issuing opposite one another and consisting 
of knots and haze, the whole affected with much irregularity 
but retaining a general symmetry. From the nature of their 
origin, the streams lie nearly in the plane of the orbit of the 
