by Carl Sagan
At this point, a little history must enter our story. In the 1930s and early 1940s, the only astronomer in the world concerning himself with planetary chemistry was the late Rupert Wildt, once of Göttingen, and later at Yale. It was Wildt who first identified methane in the atmospheres of Jupiter and Saturn, and it was he who first proposed the presence of higher hydrocarbon gases in the atmospheres of these planets. Thus, the idea that “petroleum gases” might exist on Jupiter is not original with Velikovsky. Likewise, it was Wildt who proposed that formaldehyde might be a constituent of the atmosphere of Venus, and that a carbohydrate polymer of formaldehyde might constitute the clouds. The idea of carbohydrates in the clouds of Venus was not original with Velikovsky either, and it is difficult to believe that one who so thoroughly researched the astronomical literature of the 1930s and 1940s was unaware of these papers by Wildt which relate so closely to Velikovsky’s central theme. Yet there is no mention whatever of the Jupiter phase of Wildt’s work and only a footnote on formaldehyde (page 368), without references, and without any acknowledgment that Wildt had proposed carbohydrates on Venus. Wildt, unlike Velikovsky, understood well the difference between hydrocarbons and carbohydrates; moreover, he performed unsuccessful spectroscopic searches in the near-ultraviolet for the proposed formaldehyde monomer. Being unable to find the monomer, he abandoned the hypothesis in 1942. Velikovsky did not.
As I pointed out many years ago (Sagan, 1961), the vapor pressure of simple hydrocarbons in the vicinity of the clouds of Venus should make them detectable if they comprise the clouds. They were not detectable then, and in the intervening years, despite a wide range of analytic techniques used, neither hydrocarbons nor carbohydrates have been found. These molecules have been searched for by high-resolution ground-based optical spectroscopy, including Fourier transform techniques; by ultraviolet spectroscopy from the Wisconsin Experimental Package of the Orbiting Astronomical Observatory OAO-2; by ground-based infrared observations; and by direct entry probes of the Soviet Union and the United States. Not one of them has been found. Typical abundance upper limits on the simplest hydrocarbons and on aldehydes, the building blocks of carbohydrates, are a few parts per million (Connes, et al., 1967; Owen and Sagan, 1972). [The corresponding upper limits for Mars are also a few parts per million (Owen and Sagan, 1972)]. All observations are consistent in showing that the bulk of the Venus atmosphere is composed of carbon dioxide. Indeed, because the carbon is present in such an oxidized form, at best trace constituents of the simple reduced hydrocarbons could be expected. Observations on the wings of the critical 3.5 micron region show not the slightest trace of the C-H absorption feature common to both hydrocarbons and carbohydrates (Pollack, et al., 1974). All other absorption bands in the Venus spectrum, from the ultraviolet through the infrared, are now understood; none of them is due to hydrocarbons or carbohydrates. No specific organic molecule has ever been suggested that can explain with precision the infrared spectrum of Venus as it is now known.
Moreover, the question of the composition of the Venus clouds-a major enigma for centuries-was solved not long ago (Young and Young, 1973; Sill, 1972; Young, 1973; Pollack, et al., 1974). The clouds of Venus are composed of an approximately 75 percent solution of sulfuric acid. This identification is consistent with the chemistry of the Venus atmosphere, in which hydrofluoric and hydrochloric acid have also been found; with the real part of the refractive index, deduced from polarimetry, which is known to three significant figures (1.44); with the 11.2 micron and 3 micron (and, now, far-infrared) absorption features; and with the discontinuity in the abundance of water vapor above and below the clouds. These observed features are inconsistent with the hypothesis of hydrocarbon or carbohydrate clouds.
With such organic clouds now so thoroughly discredited, why do we hear about space-vehicle research having corroborated Velikovsky’s thesis? This also requires a story. On December 14, 1962, the first successful American interplanetary spacecraft, Mariner 2, flew by Venus. Built by the Jet Propulsion Laboratory, it carried, among other more important instruments, an infrared radiometer for which I happened to be one of four experimenters. This was at a time before even the first successful lunar Ranger spacecraft, and NASA was comparatively inexperienced in releasing the scientific findings. A press conference was held in Washington to announce the results, and Dr. L. D. Kaplan, one of the experimenters on our team, was delegated to describe the results to the assembled reporters. It is clear that when his time came, he described the results with somewhat the following flavor (these are not his exact words): “Our experiment was a two-channel infrared radiometer, one channel centered in the 10.4 micron CO2 hot band, the other in an 8.4 micron clear window in the gas phase of the Venus atmosphere. The objective was to measure absolute brightness temperatures and differential transmission between the two channels. A limb-darkening law was found in which the normalized intensity varied as mu to the power alpha, where mu is the arccosine of the angle between the local planetary normal and the line of sight, and-”
At some such point he was interrupted by impatient reporters, unused to the intricacies of science, who said something like “Don’t tell us the dull stuff; give us the real poop! How thick are the clouds, how high are they, and what are they made of?” Kaplan replied, quite properly, that the infrared radiometer experiment was not designed to test such questions, nor did it. But then he said something like “I’ll tell you what I think.” He went on to describe his view that the greenhouse effect, in which an atmosphere is transparent to visible sunlight but opaque to infrared emission from the surface, needed to keep the surface of Venus hot, might not work on Venus because the atmospheric constituents seemed to be transparent at a wavelength in the vicinity of 3.5 microns. If some absorber at this wavelength existed in the Venus atmosphere, the window could be plugged, the greenhouse effect retained, and the high surface temperature accounted for. He proposed that hydrocarbons would be splendid greenhouse molecules.
Kaplan’s cautions were not noted by the press, and the next day headlines could be found in many American newspapers saying: “Hydrocarbon Clouds Found on Venus by Mariner 2.” Meanwhile, back at the Jet Propulsion Laboratory, several Laboratory publicists were in the process of writing a popular report on the mission, since called “Mariner: Mission to Venus.” One imagines them in the midst of writing, picking up the morning newspaper and saying, “Hey! I didn’t know we found hydrocarbon clouds on Venus.” And, indeed, that publication lists hydrocarbon clouds as one of the principal discoveries of Mariner 2: “At their base, the clouds are about 200 degrees F and probably are comprised of condensed hydrocarbons held in oily suspension.” (The report also opts for greenhouse heating of the Venus surface, but Velikovsky has chosen to believe only a part of what was printed.)
One now imagines the Administrator of NASA passing on the good tidings to the President in the annual report of the Space Administration; the President handing it on yet another step in his annual report to Congress; and the writers of elementary astronomy texts, always anxious to include the very latest results, enshrining this “finding” in their pages. With so many apparently reliable, high-level and mutually consistent reports that Mariner 2 found hydrocarbon clouds on Venus, it is no wonder that Velikovsky and several fair-minded scientists, inexperienced in the mysterious ways of NASA, might deduce that here is the classic test of a scientific theory: an apparently bizarre prediction, made before the observation, and then unexpectedly confirmed by experiment.
The true situation is very different, as we have seen. Neither Mariner 2 nor any subsequent investigation of the Venus atmosphere has found evidence for hydrocarbons or carbohydrates, in gas, liquid or solid phase. It is now known (Pollack, 1969) that carbon dioxide and water vapor adequately fill the 3.5 micron window. The Pioneer Venus mission in late 1978 found just the water vapor needed, along with the long-observed quantity of carbon dioxide, to account for the high surface temperature through the greenhouse effect. It is ironic that the Mariner 2
“argument” for hydrocarbon clouds on Venus in fact derives from an attempt to rescue the greenhouse explanation of the high surface temperature, which Velikovsky does not support. It is also ironic that Professor Kaplan was later a co-author of a paper that established a very low abundance of methane, a “petroleum gas,” in a spectroscopic examination of the Venus atmosphere (Connes, et al., 1967).
In summary, Velikovsky’s idea that the clouds of Venus are composed of hydrocarbons or carbohydrates is neither original nor correct. The “crucial test” fails.
PROBLEM VIII. THE TEMPERATURE OF VENUS
ANOTHER CURIOUS circumstance concerns the surface temperature of Venus. While the high temperature of Venus is often quoted as a successful prediction and a support of Velikovsky’s hypothesis, the reasoning behind his conclusion and the consequences of his arguments do not seem to be widely known nor discussed.
Let us begin by considering Velikovsky’s views on the temperature of Mars (pages 367-368). He believes that Mars, being a relatively small planet, was more severely affected in its encounters with the more massive Venus and Earth, and therefore that Mars should have a high temperature. He proposes that the mechanism may be “a conversion of motion into heat,” which is a little vague, since heat is precisely the motion of molecules or, much more fantastic, by “interplanetary electrical discharges” which “could also initiate atomic fissions with ensuing radioactivity and emission of heat.”
In the same section, he baldly states, “Mars emits more heat than it receives from the Sun,” in apparent consistency with his collision hypothesis. This statement is, however, dead wrong. The temperature of Mars has been measured repeatedly by Soviet and American spacecraft and by ground-based observers, and the temperatures of all parts of Mars are just what is calculated from the amount of sunlight absorbed by the surface. What is more, this was well known in the 1940s, before Velikovsky’s book was published. And while he mentions four prominent scientists who were involved before 1950 in measuring the temperature of Mars, he makes no reference to their work, and explicitly and erroneously states that they concluded that Mars gave off more radiation than it received from the Sun.
It is difficult to understand this set of errors, and the most generous hypothesis I can offer is that Velikovsky confused the visible part of the electromagnetic spectrum, in which sunlight heats Mars, with the infrared part of the spectrum, in which Mars largely radiates to space. But the conclusion is clear. Mars, even more than Venus, by Velikovsky’s argument should be a “hot planet.” Had Mars proved to be unexpectedly hot, perhaps we would have heard of this as a further confirmation of Velikovsky’s views. But when Mars turns out to have exactly the temperature everyone expected it to have, we do not hear of this as a refutation of Velikovsky’s views. There is a planetary double standard at work.
When we now move on to Venus, we find rather similar arguments brought into play. I find it odd that Velikovsky does not attribute the temperature of Venus to its ejection from Jupiter (see Problem I, above), but he does not. Instead, we are told, because of its close encounter with the Earth and Mars, Venus must have been heated, but also (page 77) “the head of the comet… had passed close to the Sun and was in a state of candescence.” Then, when the comet became the planet Venus, it must still have been “very hot” and have “given off heat” (page ix). Again pre-1950 astronomical observations are referred to (page 370), which show that the dark side of Venus is approximately as hot as the bright side of Venus, to the level probed by middle-infrared radiation. Here Velikovsky accurately quotes the astronomical investigators, and from their work deduces (page 371) “the night side of Venus radiates heat because Venus is hot.” Of course!
What I think Velikovsky is trying to say here is that his Venus, like his Mars, is giving off more heat than it receives from the Sun, and that the observed temperatures on both the night and day sides are due more to the “candescence” of Venus than to the radiation it now receives from the Sun. But this is a serious error. The bolometric albedo (the fraction of sunlight reflected by an object at all wavelengths) of Venus is about 0.73, entirely consistent with the observed infrared temperature of the clouds of Venus of about 240°K; that is to say, the clouds of Venus are precisely at the temperature expected on the basis of the amount of sunlight that is absorbed there.
Velikovsky proposed that both Venus and Mars give off more heat than they receive from the Sun. He is wrong for both planets. In 1949 Kuiper (see References) suggested that Jupiter gives off more heat than it receives, and subsequent observations have proved him right. But of Kuiper’s suggestion Worlds in Collision breathes not a word.
Velikovsky proposed that Venus is hot because of its encounters with Mars and the Earth, and its close passage to the Sun. Since Mars is not anomalously hot, the high surface temperature of Venus must be attributed primarily to the passage of Venus near the Sun during its cometary incarnation. But it is easy to calculate how much energy Venus would have received during its close passage to the Sun and how long it would take for this energy to be radiated away into space. This calculation is performed in Appendix 3, where we find that all of this energy is lost in a period of months to years after the close passage to the Sun, and that there is no chance of any of that heat being retained at the present time in Velikovsky’s chronology. Velikovsky does not mention how close to the Sun Venus is supposed to have passed, but a very close passage compounds the already extremely grave collision physics difficulties outlined in Appendix 1. Incidentally, there is a slight hint in Worlds in Collision that Velikovsky believes that comets shine by emitted rather than reflected light. If so, this may be the source of some of his confusion regarding Venus.
Velikovsky nowhere states the temperature he believed Venus to be at in 1950. As mentioned above, on page 77 he says vaguely that the comet that later became Venus was in a state of “candescence,” but in the preface to the 1965 edition (page xi), he claims to have predicted “an incandescent state of Venus.” This is not at all the same thing, because of the rapid cooling after its supposed solar encounter (Appendix 3). Moreover, Velikovsky himself is proposing that Venus is cooling through time, so what precisely Velikovsky meant by saying that Venus is “hot” is to some degree obscure.
Velikovsky writes in the 1965 preface that his claim of a high surface temperature was “in total disagreement with what was known in 1946.” This turns out to be not quite the case. The dominant figure of Rupert Wildt again looms over the astronomical side of Velikovsky’s hypothesis. Wildt, who, unlike Velikovsky, understood the nature of the problem, predicted correctly that Venus and not Mars would be “hot.” In a 1940 paper in the Astrophysical Journal, Wildt argued that the surface of Venus was much hotter than conventional astronomical opinion had held, because of a carbon-dioxide greenhouse effect. Carbon dioxide had recently been discovered spectroscopically in the atmosphere of Venus, and Wildt correctly pointed out that the observed large quantity of CO2 would trap infrared radiation given off by the surface of the planet until the surface temperature rose to a higher value, so that the incoming visible sunlight just balanced the outgoing infrared planetary emission. Wildt calculated that the temperature would be almost 400°K, or around the normal boiling point of water (373°K = 212 °F = 100°C). There is no doubt that this was the most careful treatment of the surface temperature of Venus prior to the 1950s, and it is again odd that Velikovsky, who seems to have read all papers on Venus and Mars published in the Astrophysical Journal in the 1920s, 1930s and 1940s, somehow overlooked this historically significant work.
We now know from ground-based radio observations and from the remarkably successful direct entry and landing probes of the Soviet Union that the surface temperature of Venus is within a few degrees of 750°K (Marov, 1972). The surface atmospheric pressure is about ninety times that at the surface of the Earth, and is comprised primarily of carbon dioxide. This large abundance of carbon dioxide, plus the smaller quantities of water vapor which have been detected on Venus,
are adequate to heat the surface to the observed temperature via the greenhouse effect. The Venera 8 descent module, the first spacecraft to land on the illuminated hemisphere of Venus, found it illuminated at the surface, and the Soviet experimenters concluded that the amount of sunlight reaching the surface and the atmospheric constitution were together adequate to drive the required radiative-convective greenhouse (Marov, et al., 1973). These results were confirmed by the Venera 9 and 10 missions, which obtained clear photographs, in sunlight, of surface rocks. Velikovsky is thus certainly mistaken when he says (page ix) “light does not penetrate the cloud cover,” and is probably mistaken when he says (page ix) the “greenhouse effect could not explain so high a temperature.” These conclusions received important additional support late in 1978 from the U.S. Pioneer Venus mission.
A repeated claim by Velikovsky is that Venus is cooling off with time. As we have seen, he attributes its high temperature to solar heating during a close solar passage. In many publications Velikovsky compares published temperature measurements of Venus, made at different times, and tries to show the desired cooling. An unbiased presentation of the microwave brightness temperatures of Venus-the only nonspacecraft data that apply to the surface temperature of the planet-are exhibited in Figure 1. The error bars represent the uncertainties in the measurement processes as estimated by the radio observers themselves. We see that there is not the faintest hint of a decline in temperature with time (if anything, there is a suggestion of an increase with time, but the error bars are sufficiently large that such a conclusion is also unsupported by the data). Similar results apply to measurements, in the infrared part of the spectrum, of cloud temperatures: they are lower in magnitude and do not decline with time. Moreover, the simplest considerations of the solution of the one-dimensional equation of heat conduction show that in the Velikovskian scenario essentially all the cooling by radiation to space would have occurred long ago. Even if Velikovsky were right about the source of the high Venus surface temperatures, his prediction of a secular temperature decrease would be erroneous.