These two bits of evidence are far too slight to form the foundation of any theory as to changes in solar activity in the geological past. Nevertheless it may be helpful to set forth certain possibilities as a stimulus to further research. For example, it has been suggested that meteoric bodies may have fallen into the sun and caused it suddenly to flare up, as it were. This is not impossible, although it does not appear to have taken place since men became advanced enough to make careful observations. Moreover, the meteorites which now fall on the earth are extremely small, the average size being computed as no larger than a grain of wheat. The largest ever found on the earth's surface, at Bacubirito in Mexico, weighs only about fifty tons, while within the rocks the evidences of meteorites are extremely scanty and insignificant. If meteorites had fallen into the sun often enough and of sufficient size to cause glacial fluctuations and historic pulsations of climate, it seems highly probable that the earth would show much more evidence of having been similarly disturbed. And even if the sun should be bombarded by large meteors the result would probably not be sudden cold periods, which are the most notable phenomena of the earth's climatic history, but sudden warm periods followed by slow cooling. Nevertheless, the disturbance of the sun by collision with meteoric matter can by no means be excluded as a possible cause of climatic variations. Allied to the preceding hypothesis is Shapley's[113] nebular hypothesis. At frequent intervals, averaging about once a year during the last thirty years, astronomers have discovered what are known as novae. These are stars which were previously faint or even invisible, but which flash suddenly into brilliancy. Often their light-giving power rises seven or eight magnitudes--a thousand-fold. In addition to the spectacular novae there are numerous irregular variables whose brilliancy changes in every ratio from a few per cent up to several magnitudes. Most of them are located in the vicinity of nebulae, as is also the case with novae. This, as well as other facts, makes it probable that all these stars are "friction variables," as Shapley calls them. Apparently as they pass through the nebulae they come in contact with its highly diffuse matter and thereby become bright much as the earth would become bright if its atmosphere were filled with millions of almost infinitesimally small meteorites. A star may also lose brilliancy if nebulous matter intervenes between it and the observer. If our sun has been subjected to any of these changes some sort of climatic effect must have been produced. In a personal communication Shapley amplifies the nebular climatic hypothesis as follows: