Indeed, the minimum observed on the frequency curve between the two maxima closely approaches the ordinate axis, i. e. , it is very small, and consequently the transition from the upper part of the curve characterizing the continental heights to the lower part, which characterizes the depth of the oceans, is insignificant: otherwise, intermediate heights would have been abundantly represented. Their virtual absence gives firm evidence to the effect that the principal formations of the Earth's surface are genetically unrelated. This is repeatedly confirmed by the abruptness of transition from continental structures to the oceans, via the steep continental slope, which is observed not only at the fringes of continents, but also surrounding any of the large islands.
These facts suffice to prove, without touching on the entire body of geological, geophysical, astronomical, and other evidence, that the oceans, in view of their comparatively young age relative to the age of the Earth as a whole (100-150 million years as compared to several billion), are neoformations which originated independently of the continents owing to surface accretion in these depressions, at a lower level than the continents. This position is consistent with the conception of an expanding Earth, advocated by the author of the present article / 2/.
To conclude the analysis of contemporary terrestrial hypsometry, we should observe that in some oceanic zones, particularly in the Pacific Ocean, there exist vast leveled portions of the bottom generally situated at a depth of 5-6km (as distinct from the broken relief of the shallower parts). This explains the third, lowest stage of the hypsographic curve.
We can make a suggestion (there is still no direct evidence for or against that this part refers to the exposed (that is, covered by a thin sedimentary layer) peridotitic shell of the Earth. This assumption becomes more plausible when we compare the general relief of the ocean with the topography of other planets, but mainly of the Moon /3/.
Until recently this problem was not even formulated in the most general of terms, but proceeding from a comparative analysis of hypsographic data for the planets we may gain insight into the historical stages through which our own planet has passed.
To plot the frequency curve of heights and depths for the Moon, the author studied the available hypsometric (physical) maps for this body. It turned
out that besides the chart compiled over 60 years ago by Franz /6/, there only exists the map of G. Schrutke-Rachtenstamm /8/. However, the latter author, who used an enormous body of data, failed in this interpreta tion—in his charts he made no allowance for the actual outlines of the "maria", which is highly essential in view of the scant data available. The last chart was therefore redrawn by the author of the present article and presented at the International Symposium on the Moon, held in December 1960 in Leningrad /3/.
A calculation of areas in each height interval on this map made it possible to plot for the Moon the frequency curve of heights and depths, and then to draw—especially for this article—the lunar hypsographic curve (Figures 3 and 4 and Table). In all plots the bottom of crater Mosting A was assigned zero level.
However, before proceeding with our analysis—mainly an analysis of the frequency curve, it being the more characteristic of the two—we should emphasize that of necessity the accuracy of the plots is low: the isolines were drawn every 2 km, Consequently, flexures of relief, even the boundaries of lunar maria and continents, are inadequately represented in our graphs.
The lunar frequency curve, in distinction from the terrestrial curve, has a single maximum (see Figure 4). A study of spectroscopic properties of lunar rocks leaves no doubt that this curve is analogous to the curve for the ocean bottom of the Earth /3/. Like the ocean bottom, the lunar formations are apparently made of basic (lunar continents) and ultrabasic (maria) rocks. While accumulation of more accurate data on terrestrial oceans will enable a trimodal curve to be plotted for the Earth in the near future, the lunar curve will eventually have two maxima. These secondary maxima, however, will not be as sharp as the present ones.
A comparative analysis of the two curves thus leads to certain general questions. Did the frequency curves always have an appearance similar to their present one? The answer, of course, is no. In particular, on the Earth in past epochs (prior to the Mesozoic) no oceanic depressions apparently existed; they formed subsequently in extension zones of the primary crust /2/. Up to that stage, only small, so-called epicontinental seas existed on the Earth*. The frequency curve corresponding to that time should only have the upper "continental" peak.