On Mars: Exploration of the Red Planet. 1958-1978

[390] When Viking touched down on the surface, weather reports started streaming their way to Earth. Martian weather was clear, cold, uniform, repetitious. Seymour L. Hess, meteorology team leader, reported on conditions at Chryse Planitia on sols 2 and 3: *
Winds in the late afternoon were again out of a generally easterly direction but southerly components appeared that had not been seen before. Once again the winds went to the southwesterly after midnight and oscillated about that direction through what appears to be two cycles. The data ended at 2:17 PM (local Martian time) with the wind from the ESE, instead of from the W as had been seen before. The maximum mean wind speed was 7.9 meters per second (18 mph) but gusts were detected reaching 14.5 meters per second (32 mph).
The minimum temperature attained just after dawn was almost the same as on the previous Sol, namely -86°C....The maximum measured temperature at 2:18PM was -33°C....This [was] 2° cooler than measured at the same time on the previous Sol.
The mean pressure was 7.83 mb, which is slightly lower than previously. It appears that pressure varies during a Sol, being about 0.1 mb higher around 2:00AM and 0.1 mb lower around 4:00 PM. 42
During the course of the Viking lander experiments, Hess and his fellow meteorologists discovered two interesting facts about Martian weather patterns. One was the extreme uniformity of the weather, presumably....


Table 54 [391]

Mars and Earth Temperatures, 21 July 1976



United Sates

Lowest temperature

- 85.5°C

- 73.0°C


(Soviet Vostok Research Station, Antartica)

(Point Barrow, Alaska)

Highest temperature

- 30.0°C



(Timimoun, Algeria)

(Needles, California)

....due to the Martian atmosphere, which is much simpler than Earth's. The Red Planet has only very, very small amounts of water vapor and no oceans-makers of extreme weather on Earth. Earth's atmospheric and surface water contribute substantially to the variability of its weather. The second discovery was the seasonal variation of pressure. When Viking first landed, its instruments detected a steady decrease in the mean pressure from day to day. But in the extended mission, the pressure at both landing sites reached its lowest value seasonally and began to rise again. The Viking meteorologists think this variation is due to the condensation of carbon dioxide on the winter cap and its release as spring comes to the northern hemisphere. This process would remove a major constituent from the atmosphere at a certain rate, changing the pressure accordingly.
At the second Viking site, 48 north, the temperatures dropped as expected during the Martian winter. Early in the mission, the minimum temperature was about -87°C, but during winter the minimum temperature at dawn was - 118°C. Frost on the surface was first observed in mid-September 1977. At the time, the second lander was recording nighttime temperatures of -113°C, and a photo of the frost was taken at -97°C. With winter, the wind speeds increased slightly, especially at the Viking 1 site, with several interruptions in what had been a regular pattern of wind....

Viking Lander's meteorology boom and sensors in deployed configuration

[392]....direction. There were several periods of northerly winds all day for several days in a row, associated with temporary drops in temperature-Martian cold fronts. Hess and his colleagues had thought winter, with the fallout of carbon dioxide, would greatly increase wind speeds and variability. There were some wind directional changes and gusts, but no noticeable changes of patterns in wind direction or speed were recorded. 43
Hardware Problems
While Hess and his meteorological colleagues began to compile weather data for the Viking 1 landing site, other experimenters were having their difficulties. First, the seismometer was nut functioning. Its seismic sensor coils had been "caged" mechanically to prevent damage to these sensitive components during the shock of landing. Following touchdown, a fusible pin-pulling device was to have detonated, unlocking the seismometer so it could begin full operation. For some reason, perhaps a broken or misconnected wire, the fusible device failed to work, and the instrument remained in the caged position. While the Viking 2 seismometer performed satisfactorily, the Viking 1 failure prevented the seismology team from locating the approximate origin points of recorded seismic activity. 44
Don Anderson and his colleagues on the seismology team was afraid that the sensitive seismometer on Viking would be hampered by the high winds on Mars. But during the night from about 6 p.m. through the next morning, the winds die down to about virtually zero and there are essentially no seismic background noises. During that time, the seismometer can be operated "at a very high sensitivity." Marsquakes as small as a magnitude of 3 at a distance of about 200 kilometers can be recorded. By comparing a Marsquake with a similar Earthquake, the specialists estimated the mean crustal thickness at the Viking 2 landing site to be about 14 to 18 km, about half the thickness of the crust in the continental parts of Earth and about 50 percent greater than the average thickness of the oceanic crust. Viking scientists think the crust on Mars may be as thick as approximately 80 km, much thicker than the crust under continental regions on Earth.
An unexpected result of the seismic experiment was a great amount of information about the winds on Mars. A very sensitive wind detector, the seismometer picks up the wind pressure on the lander, from which characteristics of the wind can be determined. Like the meteorologists, the seismology team detected the cold fronts. The wind pattern "changed very rapidly on the 131st Martian day. The winds....started to blow all night until 2 or 3 a.m. indicating a substantial change in the weather patterns. If very high winter winds had continued at night, they could have generated the massive dust storms we have observed in the winter time." However, orbiter photographs have shown only a few isolated dust storms, with none reaching the magnitude of the planetwide dust storm of 1971. 45
[393] Another cause for concern for the Viking team appeared on the second day of landed operations. The lander's UHF transmitter had been designed to operate at three different power levels-1 watt, 10 watts, and 30 watts- depending on the rate of data transmission required. During the relay-link portions of the mission, the 30-watt power level was scheduled for use, to permit the transmission of the maximum amount of scientific data. From the observed performance of the initial lauded relay link, confidence in the system was high. During the first relay, approximately 30 million bits of data were transmitted to the orbiter, recorded, and subsequently transmitted to Earth, all within a few hours after the information left the surface of Mars.** Success however, was short-lived.
On 22 and 23 July, the UHF transmitter switched over to the 1-watt power level without instructions to do so. Tom Young told the press, "In the one-watt mode you can get slightly over seventeen minutes' worth of data from the Lander to the Orbiter." The mission had been designed so that slightly more than l8 minutes of data would be transmitted to the orbiter as it passed overhead, so the problem was not critical one, but it did pose a vexing limitation. At the 30-watt level, the lander could transmit telemetry to the orbiter for 30 to 32 minutes. 46
On the morning of 24 July, the UHF transmitter switched back to the 30-watt power level. Tom Young reported this second mysterious power change at the news briefing that day: "When we had the relay [of information] today, lo and behold, it came up in the 30-watt mode, operating as we would like for it to. So our statistics, to date, are two relay periods in the 1-watt mode, two periods in the 30-watt mode. We are continuing the analysis of this particular anomaly. 47 The radio specialists suspected that the problem lay in the power-mode control-logic subassembly of the UHF transmitter. To counteract this trouble, commands had been prepared to order the guidance control and sequencing computer to eliminate the electronic "noise" causing the problem. Before this command was sent up, the transmitter switched back to the 30-watt power level. The change supported the theory that the problem was associated with noise susceptibility. Following the self-correction, the UHF transmitter performed as expected until one week before the end of Viking 1 's primary mission. At that time, telemetry indicated that there were potentially new problems with the 30-watt level. To avoid a catastrophic failure and to extend the transmitter's life for use in the "follow-on'' mission, the lander performance analysts decided to use the 10-watt power mode for the last sols of the basic mission. 48
The landed relay communications for Viking 2 did not demonstrate ally anomalies. On sol 21 of the second landed mission, orbiter I was moved into position over lander 2 to provides relay link. This maneuver permitted mission planners to send orbiter 2 on an extended "walk" around the planet, to photograph the poles and other regions of Mars and scan them [394] with the infrared thermal mapper and the Martian atmospheric water detector. Orbiter l continued to provide the communications link for the second lander during the remainder of Viking 2 's primary mission. 49
A more serious problem emerged in the first days after Viking 1 's touchdown when the surface sampler arm became stuck. On Thursday, 22 July, the surface sampler assembly was rotated so that the protective shroud covering the sample collector head (scoop) could be jettisoned. During this operation, the sampler boom was to be extended a few centimeters and then returned to the stowed position. Extending the boom was no problem, but on retraction it stuck. At first, Jim Martin and crew thought the problem was one of electronics. At 6:30 p.m. on the 22d, Martin told reporters preliminary indications were that perhaps the soil-sampler control assembly-the receiver for computer commands-had "some kind of an electronic problem." He could switch to a redundant soil-sampler control assembly if that was the problem, but, "the concern I have at the moment is that unless we can solve or understand this problem and solve it in fairly short order we are likely to run the risk of impacting the soil acquisition sequence on Sol 8." 50
By 10 p.m. on the 22d, Martin's team had arrived at a new theory. Prefacing his remarks to the media with, "It has been a very busy day,'' Martin addressed the problem of the sampler. Everyone knew that loss of the sampler would be a major setback for Viking science activities. Without It, no samples would be delivered to the biology instrument, the gas chromatograph-mass spectrometer, or the x-ray fluorescence spectrometer. Martin believed that his people, who had worked all evening, had "isolated the most probable cause of the problem. It turns out, contrary to my expectation, not to beau electrical problem. "Instead, it was apparently a simple- if anything can be simple when working with a piece of equipment millions of kilometers away-mechanical hang-up. Martin pointed out "that there is a locking pin that is part of the shroud latching system"; that pin was supposed to drop to the Martian surface doting the boom extension... It now appears that the extension that had been commanded in the sequence was not long enough to allow this pin to drop free."
Martin bad observed a duplication of the difficulty on the science test lander, which was housed in a glass-walled room next to the auditorium in which the press briefings were held at JPL. Commenting on the fishbowl atmosphere in which his people had been working, Martin told the reporters, "I went in and looked at it myself when some of you weren't looking.'' The stuck pin was "certainly a plausible and possible failure mode." To test this theory, "we plan to send up a new command sequence on the Sol 5 command load which will go up at around midnight Saturday night," 24 July. Mission analysts thought that extending the boom to about 35 centimeters would let the pin fall. Martin added, "If by some chance the pin was retained within the mechanism, which really believe is doubtful, we don't ever intend to retract it as far as we did in the original sequence." That way. they would avoid another difficulty; at a certain point the boom extraction motor would clutch on purpose and then shut itself off to avoid [395] damage to the motor. If the pin did not drop free this time, the boom would be ordered to extend far enough so that the "no-go'' signal would not be given. 51
Two photographs taken by the lander camera on sol 5, 25 July, showed that the retaining pin did fall free, landing on the ground in front of the craft. 52 The apparent ease with which this problem had been diagnosed and corrected hid the months of training and preparation fur such mission operations. Subsequently, more serious troubles were to plague the soil- sampler assembly, but each time training and ingenuity permitted the team to work out solutions and keep the mechanism functioning. Adaptability was one of the key elements of Viking's landed operations.
Communicating with the Spacecraft
Before separation from the orbiter, the lander had been given an initial computer load (ICL, or "ickel"), which contained all the computer commands necessary for a basic 60-day mission, even if there were no further communications from Earth. With normal communications between the spacecraft and mission control, the mission programmers could modify the initial computer load as needed to get the most out of the lander. Commands were "uplinked" to the lander from JPL through the stations of the Deep Space Network to the orbiter and then to the guidance, control, and sequencing computer. The command uplinks, made in three-day cycles, were the responsibility of the lander command and sequencing team of the lander performance and analysis group.
Agreeing on the commands to be sent to the lander, programming them, and checking them out through simulations was a complex series of tasks, which required a great deal of work and interaction among many persons. An example is the decision to photograph the sampler boom immediately after acquisition of a sample. The requirement would first be sent to the lander imaging team, which had three three-person squads who handled such requests. These uplink squads, plus a "late-adaptive squad" responsible for last-minute alterations, would investigate the picture called for and determine if it could be combined with others or if it had to be taken by itself. The series of pictures for a given sol was then described and combined into a science requirement strategy that was passed on to the Lander Science Systems Staff, which had the difficult task of matching wants (requirements) with the constraints imposed by the lander systems and the other tasks that had to be accomplished.
The Lander Science Systems Staff received the uplink plans in the form of computer printouts called science instrument parameters-specific commands to the guidance, control, and sequencing computer. Lander Imaging had 56 commands available, and each could be adapted to special requirements. Once approved by the Lander Science Systems Staff, the parameters were passed on to the lander computer simulations personnel, who ran through the commands to see if there were any software or hardware...
[396] Viking Surface Sampler
The Viking lander's chemical and biological investigations all used samples of surface materials excavated by the surface sampler In addition, as the experience with lunar Surveyor spacecraft demonstrated, there was much to learn about the surface simply by digging in it. In the Viking mission, digging was part of the physical properties and magnetic properties investigations.
The surface sampler consisted of a col rector head attached to the end of a three-meter retractable boom. The arm housing the boom could be moved both horizontally and vertically. The boom itself was constructed from two ribbons of stainless steel welded together along the edges. When extended, the two layers opened to form a rigid tube. When retracted, the boom flattened. A flat cable sandwiched between the boom layers transmitted electrical power to the collector head.
The collector head was basically a scoop with a movable lid and a backhoe hinged to its lower surface. Where the scoop is attached to the end of the boom, a motorized rotator acted as a mechanical wrist to permit manipulation of the collector head. To fill the scoop, the lid was first raised and then the boom was extended along or into the surface. Once full, the lid closed. The top of the lid had holes two millimeters in diameter, which formed a sieve. When the collector head was positioned over one of the inlets for the instruments, it was inverted and vibrated. Only particles smaller than two millimeters were delivered to the instrument inlets. Coarser samples could be delivered to the x-ray fluorescence spectrometer, if desired. The gas chromatograph-mass spectrometer and the biologyinstruments had their own filters to control the size of material introduced into their sample processing assemblies.
The surface sampler could also dig trenches, by lowering the backhoe to place the sampler head on the surface, and then retracting the boom. Excavated materials could be scooped up for sampling. A brush, magnets, temperature sensor, and other instrumentation also provide data concerning the physical properties of the materials.

....conflicts. Considerations such as electrical energy required or the thermal impact of a command were also determined. Following simulations, the request was codified into a "lander sequence." After all the necessary changes (massaging) were completed, the command was entered into the ground-based computer and relayed to the Deep Space Network for transmission to Mars.

Uplink teams preparing lander sequences worked about two weeks ahead of the time the command was to he executed. Changes could be made in the planned uplink until about 48 hours before it was loaded into the....

[Whole page 397]

A premission photo, above, shows how the surface-sampler collector head deposits its contents into the biology-instrument processor and distributor assembly. The collector head and the area of the sampler arm's operation are sketched at right. Below, project scientist Gerals Soffen examines the collector head on the science test lander at JPL.


[398]....computer. Obviously, uplinking was a precise. demanding business. Mistakes were totally inadmissible. Although out of the limelight, the people responsible for talking with the lander had a difficult task. Occasionally, nerves wore thin when the requirements of different science teams conflicted. The uplinkers were expected to satisfy everyone's needs. and for the most part they did. 53
Sampling the Martian Surface
Scientifically, the most important experiments aboard the lander were those which sampled the planet's surface. Of these, the chemical analyses were interesting, but the biological experiments were a disappointment. As with other investigations, Mars again turned out to be a more complex riddle than anticipated and, while there is still disagreement over the exact causes of some of the reactions observed. most-but not all-of the Viking scientists have come to the opinion that detection of life on Mars is a very unlikely prospect.
The first soil samples were acquired on sol 8, 28 July. Four samples were dug, with the first being deposited into the biology instrument distributor assembly, the next two into the GCMS processor, and the fourth into the funnel of the x-ray fluorescence spectrometer. All the commands were successfully executed, but there was no positive indication that the gas chromatograph-mass spectrometer processor had been properly filled. A second acquisition attempt still did not provide a "sample level detector `full' indication." The sampler system, having completed its programmed sequences in a normal manner, parked the boom as planned. On Earth, the lander performance specialists began to analyze the possible causes of the anomaly: (1) insufficient sample acquired in the collector head because the same sample collection site had also been used for the biology sample; (2) insufficient time allowed for the sample to pass from the funnel through the sample grinding section and then through the fine (300-micrometer) sieve into the metering cavity of the instrument; (3) grinder stirring spring not contacting the sieve; or (4) sample-level-detector circuit faulty. Since the "level-full" detector consisted of a very fine wire stretched across the cavity to which the sample material was delivered, it was also possible that it had broken when the soil was dropped into the funnel. 54
An anomaly team headed by Joseph C. Moorman, who had worked closely with the builders of the GCMS, went to work on this problem. While preparations were made for another sample to be collected on sol 14, 3 August, Martin and Young had to decide whether to proceed on the assumption that the GCMS had actually been filled and chance wasting one of the two remaining ovens on an empty chamber (the specialists had determined that one of the ovens was inoperable during the GCMS in-flight checkout) or pick up another sample on sol 14. Conservatism and caution argued for the latter decision, and the managers chose that option. But the boom did not cooperate. It jammed.
[399] The surface-sampler control-assembly sequences performed normally through the 12th command. During the execution of the 13th (boom retraction to 26.7 centimeters). trouble showed up; when the computer issued the 14th command, the assembly would not respond. Examination of photos taken on sol 14 revealed that the sampling trench had been dug as ordered, but the collector head was not over the GCMS funnel where it was supposed to be. An image received on sol 15 showed the back of the boom. Three possible reasons for this new anomaly were considered: (l) failure of the surface-sampler control-assembly electronics; (2) failure of the boom motor or related equipment; or (3) jamming of the boom, precluding proper retraction. Causes 1 and 2 were rejected after analyzing the proper performance through the first 12 commands. Jamming had most likely caused the difficulty since the failure appeared to be similar to the "no-go'' response encountered with the sol 2 shroud-pin jam.
Frozen carbon dioxide or surface material were rejected as possible causes of jamming the boom, because of the absence of a slowly increasing motor load, which the investigators would have detected. Discussion of the anomaly with the boom designers revealed that a similar problem had occurred during early test phases, and they believed it was caused when a series of successive retract (or extend) commands had been issued. In testing, the successive commands tightened the boom element on the storage drum, and the boom element tended to wind around the drum in a 5-or 6-sided configuration rather than in a perfect circle. This arrangement caused Intermittent high loading when the "points of the hexagon" passed under the boom restraint brake shoes. The reliability of the system was further weakened when operated at low temperatures; the motor torque limiter finally decoupled, and movement of the boom ceased. Two major operating procedures were proposed to meet the problem: (I) All sequences were to be revised to eliminate successive extend or retract commands. avoiding excessive tightening of the boom element on the drum. The command reversals would cause the extend or retract "flip-flop" gear to disengage the load during each cycle, allowing the motor to attain full speed and operating torque before it reengaged the load in the opposite direction. (2) Future operations were to be performed within one to two hours of the peak temperature during the Martian sol. An uplink diagnostic sequence was designed for sol 18; the boom would be used in each axis of operation- extend, retract, up elevation, down elevation, clockwise, and counter-clockwise. The sequence was executed properly and no anomalies were met. Following Martin Marietta's instructions, all activities of the sampler arm were redesigned "to exclude, wherever possible. successive extend or retract commands, and to perform these operations during the warmest part of the sol." The Viking team had no further problems with the sampler boom on either lander, and operating temperature restrictions were eventually waived because of the need to acquire early morning biology samples. Preflight testing and the documentation of those procedures had paid off. 55
[400] The sol 14 anomaly forced Martin and Young to reconsider their decision not to analyze the "possible" sample acquired on sol 8. Influenced by early results from the biology experiments, the molecular analysis team urged that the contents of the gas chromatograph-mass spectrometer be analyzed. Jim Martin and Tom Young agreed.
Biology . At the 1:30 pm news briefing on 31 July 1976 (sol 11), Jim Martin made an announcement. Prefacing his remarks with. "I wanted to state that it's been project policy for seven years to make data available to the media when we have [them]," Martin noted that this day was -no exception. We have received biology data that we believe to be good data." Engineering telemetry indicated that the biology instrument was performing "extremely well," perhaps too well, since early reactions from the gas-exchange and labeled-release experiments were very positive. That could possibly be the consequence of biological activity, but Martin was cautious: "I think Chuck Klein will continue to caution you that the biology experiment is a complex one. We've seen that Mars is a complex planet. There are many things that we do not understand." The scientists were proceeding systematically and methodically. 56
Biology Team Leader Harold P. Klein and his colleagues had already conducted a number of tutorials for the news people covering the Viking mission, and at each session where they presented analytical details they took time to explain the experiment in question. The biologists started with the basics. Each Viking Lander carried an integrated biology instrument, which contained three experiments designed to detect the metabolic activity of microorganisms should they be present in the soil sampled. First, the gas-exchange experiment would determine if changes caused by microbial metabolism occurred in the composition of the test chamber atmosphere. Second, the labeled-release experiment, also known as Gulliver, would determine if decomposed organic compounds were produced by microbes when a nutrient was added. Third, the pyrolytic-release experiment would detect, from gases in the chamber, any synthesis of organic matter in the Martian soil. A change could be the result of either photosynthetic or nonphotosynthetic processes.
On 31 July, Klein told the press: "What we are proposing to do for you today [is] to give you a status report on the three experiments and we'd like to then focus on one of the experiments, the labeled release experiment, a little more closely since some of that data is exciting and interesting." First, all three instruments were working normally. "We have no anomalies, no problems despite what some of the press or other news media have said." He had heard rumors that the biology instrument was "sick, dead in the water." The truth was that the instrument was in good shape, and he had two important, unique facts.
First, the gas-exchange experiment had given them reason to believe that "we have at least preliminary evidence for a very active surface material. . . . .We believe that there's something in the surface, some chemical or [401] physical entity which is affording the surface material a great activity." But, adding a word of caution, he noted that the reaction observed in the gas-exchange experiment might be mimicking some aspects of biological activity. Second, the labeled-release experiment's radioactivity counters were measuring "a fairly high level of radioactivity which to a first approximation would look very much like a biological signal." The highly active nature of the soil, however, caused the biology team members to be cautious. "That second result must be viewed very, very carefully in order to be certain that we are, in fact, dealing with a biological or non-biological" phenomenon.
Klein reported on the sequencing of the three biology experiments. Norman Horowitz's pyrolytic-release experiment had been started first. After the soil had been injected into the test chamber and carbon 14- labeled carbon dioxide added, the xenon lamp had been turned on; incubation would last until at least sol 14, when the first results might be available. Vance Oyama's gas-exchange experiment had also received its soil sample on 28 July, but the incubation process was not begun until the morning of the 29th, when the chamber containing the soil and Martian atmosphere was injected with a mixture of carbon dioxide, krypton, and half a cubic centimeter of nutrient. About two hours later, gas in the chamber was analyzed-a calibrating measurement against which all subsequent analyses would be measured. Calling for the lights in the Von Karman Auditorium to be turned off, Klein had a chromatogram based on the first gas exchange results projected on the screen behind him:

Biology instrument

[402] What we saw were five peaks-little tiny peaks: neon, over here on your left and that's explainable by the neon we used in the nutrient chamber itself and that's our indication that we, in fact, injected nutrient and that's fine- there's nothing unusual about that. Then you see nitrogen and that amount of nitrogen can be accounted for by the nitrogen in the atmosphere and a small amount of nitrogen that we know was contaminating our CO2 krypton mixture. Then we see this oxygen peak which I will come back to in a moment. And then as a shoulder beside the oxygen, you see a small peak and that's a combination of argon and carbon monoxide and that amount of gas would be consistent with current estimates of argon and carbon monoxide in the atmosphere.
A large krypton peak, Klein explained, was present because they had added krypton in a specifically known amount to provide a standard reference for determining the amount of other gases that might be present. He turned back to the oxygen peak: "You will see at the base of that oxygen peak, a little bar-that's the amount of oxygen down there that we can account for, or could account for from all known sources in the atmosphere or in the contamination of our gas mixture," But the instrument on Viking 1 was indicating 15 times more oxygen than the scientists could account for from known sources, The results from the second measurement made 24 hours later showed that all the gases had remained the same except oxygen. It had increased by 30 percent. After ruling out all other possible causes, the scientists concluded that the oxygen had to be coming from the soil itself. While one possible explanation for the increase was biological activity, other explanations were possible, too. 57
A possible alternative answer to why the initial amount of oxygen had been released lay in the desert area of landing site; the Martian samples contained peroxides and superoxides, which when exposed to abnormal (non-Marslike) humidity in the instrument quickly released oxygen. The related release of carbon dioxide suggested that the samples had an alkaline core. Although such reactions had not been witnessed on Earth, the scientists believed that the intense ultraviolet radiation bombarding the surface of the Red Planet could have produced unique photocatalytic effects. Still, there was much to be explained, including the reactions observed from the labeled-release investigation.
Gulliver was sending back some surprises. As with the gas-exchange experiment, the labeled-release experiment added a small amount of nutrient to the soil sample. It also produced a large amount of gas after that injection. Where the gas-exchange produced a spectrum of the gases, the labeled release measured the amount of radioactivity produced by the carbon- 14 -labeling" material in the nutrient. Shortly after the addition of the nutrient, the radiation counts rose sharply, leveling off at about 10,000 counts per minute.
Gil Levin gave the audience at JPL a brief resume of the activities since the injection of the nutrients, which had occurred at about 1:45 p.m. PDT on [403] 30 July. That injection had consisted of about 0.1 milliliter, or about 2 drops, of liquid. As Levin noted. "If any organisms are present that can utilize the nutrient and if these organisms behave biochemically-roughly as terrestrial organisms do-they should imbibe the nutrient and exhale a radioactive gas." Resulting radioactivity was measured periodically by a radiation detector. The result on Mars was very interesting. It was similar to ones encountered with living organisms detected in terrestrial soil, but Levin warned, "We are far too early in the game to say that we have a positive response." There were too many factors that had to be weighed and tested. "All we can say at this point is that the response is very interesting, be it biological or non-biological, it is unanticipated."
As in the gas-exchange experiment, there was a possibility that the soil itself contained catalysts, minerals, inorganics that produced some breakdown of the radioactive compounds. "The effect of water introduced into the dry Mars soil may cause violent chemical reactions that would disintegrate a portion of our medium." As a consequence, Levin thought that any speculation about the biological or non-biological nature of the response would have to await further data. 58
By l August, the production of oxygen in the gas-exchange experiment had decreased considerably, thus supporting the belief that the release was the function of oxides in the soil. In a 2 August update on the labeled-release experiment, Levin noted that they had examined the radioactivity curve very carefully. They had found no evidence of any doubling of cells. No growth appeared to be taking place, but the curve did not seem to behave as scientists would have expected it to for chemical reactions either. "We find that the chemical reaction took place at a very rapid rate initially, and then uncharacteristically slowed down and took a long time to plateau." The curve detected with the labeled-release experiment did not agree with known responses for either chemical or biological reactions. 59
Data returned by the pyrolytic-release experiment and reported by Norman Horowitz on 7 August were equally confounding. Once again, the specialists had detected a reaction, but they did not know what it meant. "There's a possibility that this is biological," Horowitz said, but "there are many other possibilities that have to be excluded." The results obtained the night before were interesting but he emphasized that they were not ready to say that they had discovered life on Mars. "The data point we have is conceivably of biological origin, but the biological explanation is only one of a number of alternative explanations." He told the press:
We hope by the end of this mission to have excluded all but one of the explanations, whichever that may be. I want to emphasize that if this were normal science, we wouldn't even be here-we'd be working in our laboratories for three more mouths-you wouldn't even know what was going on and at the end of that time we world come out and tell you the answer. Having to work in a fishbowl like this is an experience that none of us is used to.
[404] He also cautioned the reporters that they were being included in the analysis phase of the experiments. They were "looking over the shoulder of a group of people who are trying to work in a normal way in an abnormal environment." 60 The scientist's caution was prompted by his knowledge that "we well might be wrong in anything we say. Anyone who has carried out a scientific investigation knows that the pathway of science is paved not only with brilliant insights and great discoveries, but also with false leads and bitter disappointments. And nobody wanted to be wrong in public on a question as important as that of life on Mars." 61
Later in a November 1977 Scientific American article, Horowitz was able to speak more authoritatively about the results that had been observed in all three experiments. In the gas-exchange experiment, "the findings of the first stage of the experiment were both surprising and simple." Immediately following the addition of the moisture to the sample chamber-the soil sample was not directly wetted-carbon dioxide and oxygen were released. The evolution of gases was short-lived, but the pressure in the chamber increased measurably. At the Chryse site, the amount of carbon dioxide increased by about 5 times, and the amount of oxygen increased by about 200 times in little more than one sol. At the landing site in Utopia, the increases were smaller but still "considerable." Upon reflection, Horowitz stated that "the rapidity and brevity of the response recorded by both landers suggested that the process observed was a chemical reaction, not a biological one." Horowitz felt that the appearance of the carbon dioxide was readily explainable: "Carbon dioxide gas would be expected to be adsorbed on the surface of the dry Martian soil; if the soil was exposed to very humid atmosphere, the gas would be displaced by water vapor." The presence of the oxygen was logical but harder to account for, since so much oxygen would seem to require an oxygen-producing substance, not just the physical release of preexisting gas. There was just not that much oxygen available in the atmosphere-past or present-to account for the quantities measured. Horowitz argued that it was "likely that the oxygen was released when the water vapor decomposed an oxygen-rich compound such as a peroxide. Peroxides are known to decompose if they are exposed to water in the presence of iron compounds, and according to the X-ray fluorescence spectrometer....the Martian soil is 13 percent iron."
At both sites, the second phase of the gas-exchange experiment was "anticlimactic." When the sample was saturated with the aqueous nutrient, more carbon dioxide and oxygen were produced. The additional evolution of carbon dioxide was probably a continuation of the reaction observed in the humid stage of the experiment. Horowitz believed that the amount of oxygen then diminished because of its combination with the ascorbic acid in the nutrient medium. "And so....it became clear that everything of interest happened in the humid stage of the experiment, before the soil came in contact with the nutrient!" Thus, in November 1977, Horowitz confidently stated that the gas-exchange experiment had detected "not [405] metabolism but the chemical interaction of the Martian surface material with water vapor at a pressure that has not been reached on Mars for many millions of years." 62
In the labeled-release experiment, there was a similar rapid surge of gas into the test chamber when the nutrient solution was added to the soil. This release tapered off shortly after the passage of one sol. Horowitz noted, "The gas, undoubtedly carbon dioxide, was radioactive, showing that it had been formed from the radioactive compounds of the medium and not from compounds in the Martian soil." He also believed that other nonradioactive gases were evolved when the water in the nutrient medium came in contact with the sample, but that these could not be detected by the instrument. "The production of radioactive carbon dioxide in the labeled-release experiment is understandable in light of the evidence from the gas-exchange experiment suggesting that the surface material of Mars contains peroxides." Formic acid, which was one of the compounds in the labeled-release nutrient, is oxidized with relative ease. "If a molecule of formic acid (HCOOH) reacts with one of hydrogen peroxide (H2O2), it will form a molecule of carbon dioxide (C02) and two molecules of water (H20)." The amount of radioactive carbon dioxide produced in the experiment was only slightly less than would have been predicted if all the formic acid in the nutrient had been oxidized in this manner.
Going a step further with his analysis, Horowitz said that if the source of the oxygen in the gas-exchange experiment was peroxides in the soil decomposed by the water vapor, then the labeled-release experiment should have decomposed all of the peroxides with the first injection of nutrient, The second injection should have produced no additional radioactive gas. That was what happened. "When a second volume of medium was injected into the chamber, the amount of gas in the chamber was not increased; indeed, it decreased. The decrease is explained by the fact that carbon dioxide is quite soluble in water; when fresh nutrient medium was added to the chamber, it absorbed some of the carbon dioxide in the head space above the sample."
In the labeled-release experiment, the stability of the reaction to heating at various temperatures was examined. Heating reduced and subsequently stopped the reaction. This result has been interpreted by some to be evidence in favor of biological activity, but Horowitz, although conceding that the effects of heating could be explained by biological activity, said that these results were also consistent with a chemical oxidation in which the oxidizing agent is destroyed or evaporated at relatively low temperatures. "A variety of both inorganic peroxides and organic peroxides could probably have produced the same results." 63
The third biology experiment, pyrolytic release, differed from the others in two basic respects. First, it attempted to measure the synthesis of organic matter from atmospheric gases rather than the decomposition of that matter. Second, it was designed to operate under pressure, temperature, [406] and atmospheric composition that were nearly the same as those on the planet. During the actual operation of the pyrolytic-release investigation, the temperatures ran higher than those normally encountered on Mars because of heat generated within the lander. A sample of the soil was sealed in the test chamber along with some of the planet's atmosphere. A xenon arc lamp simulated the sun. Into this Martian microcosm, small amounts of radioactive carbon dioxide and carbon monoxide were introduced. After five days, the xenon lamp was turned off, and the atmosphere was removed. The soil was then analyzed for the presence of radioactive organic matter.
Analysis of the soil began with heating it in the pyrolyzing furnace- hence, the name pyrolytic release-to a temperature high enough to reduce any organic compounds to small volatile fragments. Those "fragments were swept out of the chamber by a stream of helium and passed through a column that was designed to trap organic molecules but allow carbon dioxide and carbon monoxide to pass through." In this process, radioactive organic molecules would be transferred from the soil to the column while being separated from the remaining gases of the incubation atmosphere. Any organic molecules would be released from the column by raising the column's temperature. Simultaneously, the radioactive organic molecules would be decomposed into radioactive carbon dioxide by copper oxide in the column and transported to the radiation counter by the helium carrier gas. If, as a result of this process, organic compounds had been formed, there would be detectable radioactivity; if there were no organics, there would be no radioactivity.
Horowitz noted that, surprisingly, "seven of the nine pyrolytic-release tests executed on Mars gave positive results." The negative results occurred with samples obtained at Viking 2 's Utopia site. The amount of radioactive carbon dioxide obtained by the experiment was small; still, it was enough to furnish organic matter for between 100 and 1000 bacterial cells. Significantly, "the quantity is so small....that it could not have been detected by the organic-analysis experiment," the gas chromatograph-mass spectrometer (see below). Though small, the quantity was important, because as Horowitz expressed it, "it was surprising that in such a strongly oxidizing environment even a small amount of organic material could be fixed in the soil." Even more important to him was the fact that "the pyrolytic-release instrument had been rigorously designed to eliminate non-biological sources of organic compounds." To encounter positive results from the Martian soil in spite of all the precautions was in the biologist's word "startling."
However, on reflection, it appeared that the findings of the pyrolytic-release experiment had to be interpreted non-biologically. The reaction did not respond to heat in a manner consistent with a biological reaction. Martian microbes, accustomed to the very low temperatures on that planet, would have been killed by the elevated temperatures experienced during the test, the investigators thought. "On the other hand, it is not easy to point to a non-biological explanation for the positive results." Investigations into [407] this curious reaction have continues in terrestrial laboratories, and until "the mystery of the results. . . .is solved, a biological explanation will continue to be a remote possibility." 64
Gas Chromatograph-Mass Spectrometer (GCMS) . While the results of the biology experiments did not seem as bleak in the summer of 1976 as they have appeared subsequently, there was considerable concern during the missions about the proper interpretation of the reactions being witnessed. During August 1976, the Viking scientists believed that the GCMS was one possible tool for deciding if the reactions observed in the biology instrument were biological or chemical in origin.
As one observer noted, the gas chromatograph-mass spectrometer was the court of appeals in the event that the biological experiments did not present a clear verdict. 65 With the initial uncertainties from the biology experiments, the molecular analysis team decided to gamble that the GCMS had received its sample on sol 8 (see pages 398-400) and made the first analysis on 6 August (sol l7). Klaus Biemann reported to the press on the molecular analysis-"the first half of the first sample experiment of the organic analysis"-the following day. The soil sample was there! And the oven had worked as planned. There was always speculation among the news representatives about what new hardware problems might appear, but this time the scientists could report, "It did work as predicted, heated to 200° and stayed there for thirty seconds. The entire gas chromatograph mass spectrometer worked well like all gas chromatograph mass spectrometers do." Although the molecular analysis team was obviously pleased that its instrument was working well, the results from the GCMS would be the source of the most frustrating data for those exobiologists who were hoping to find life on the Red Planet.
About 300 mass spectra, electronically provided graphs identifying the molecules detected in the Martian soil sample, were returned by the first run of the GCMS. The molecular analysis specialists were particularly interested in determining if carbon compounds were in the sample, since biochemistry is largely the chemistry of carbon. The basic structure of the carbon atom enables it to form large and complex molecules that are very stable at ordinary temperatures. While no carbon compounds were detected in the first sample analysis, there was no great concern, since it was believed that the sample would have to be heated to 500°C before the organics would be broken down and detected by the instrument. The only surprising aspect of the first data was the very small amount of water released by the sample. 66
On 12 August, the GCMS experiment was run again with the first sample being heated to a maximum temperature of 500°C. Biemann reported that this analysis "to our surprise, evolved a large amount of water. Indeed so much that it gives us trouble in analyzing the data." Still, the critical point of this analysis was that there were probably no organics. If the reactions observed in the biology instrument were the consequence of life, then it was expected that the GCMS would detect organic compounds [408] in the same soil. Neither this analysis nor the subsequent one at the Viking 1 site, nor those carried out at the Viking 2 landing area, produced traces of organic compounds at the detection limits (a few parts per billion) of the GCMS. 67
Failure of the gas chromatograph-mass spectrometer to detect organic compounds was devastating for those who believed that life on Mars was possible. For Jerry Soffen, the GCMS results were "a real wipe out." Once he assimilated the fact that the GCMS had found no organic materials, he walked away from where the data were being analyzed saying to himself, "That's the ball game. No organics on Mars, no life on Mars, "But Soffen confessed that it took him some time to believe the results were conclusive. At first, he argued with Tom Young that there must have been no sample present in the GCMS, because there had to be organics of some sort on the planet. Soffen bet Young a dollar that the second analysis would prove that the instrument had been empty. To his dismay, the data indicated instead that there was a sample in the instrument and that the sample was devoid of organics.
Klaus Biemann, the molecular analysis team leader, had some reflections on the search for organic compounds. Looking in the soil for compounds made of carbon, hydrogen, nitrogen, and oxygen at the level of a few parts per billion, they found none. The gas chromatograph-mass spectrometer could have detected smaller concentrations of organic materials than are present in typical antarctic soil, which is low in organic compounds because there is little vegetation and animal life on that part of Earth. Compared to Antarctica, Mars is devoid of organic material, and a number of conclusions could be drawn from that finding. First, no synthesis of organic compounds is occurring on the surface, at least where the two Vikings landed. Second, if millions of years ago organic compounds did exist, they must have since been destroyed. Third, since organic compounds must be arriving on Mars in the form of meteorites, that material must have been imbedded in the surface very deeply or, more likely, destroyed by the planet's harsh environment. Finally, says Biemann, "if we use terrestrial analogies, we always find that a large amount of organic material accompanies living things-a hundred times, thousand times, 10 thousand times more organic materials than the cells themselves represent." Since the Viking instruments did not detect any large amounts of organic waste material, it is difficult to see how microorganisms could be living at the areas investigated "if they behave as terrestrial organisms do."
Of course, reminded Biemann, "this does not rule out a different kind of living mechanism that would protect its organic constituents very well and, therefore, avoid this waste of a scarce commodity." Martian organisms could have evolved along those lines, and as the environment became harsher and harsher they could have become more and more efficient in using the organic materials they needed. Viking looked at only two samples at each of the two landing sites from depths of 5 to 10 centimeters. If organic materials were produced millions or hundreds of millions of years ago, they [409] could be present at greater depths and protected there from the damaging ultraviolet radiation. The Viking spacecraft could be sitting on an area containing a deposit of organic material a few meters down. There could also be other areas on the planet where the surface material is more protected or where organic material is now being synthesized and not destroyed. To help answer these puzzling questions, Biemann and his colleagues had plans to study in their laboratories the rate of decomposition of certain typical organics under Martianlike conditions, to determine how fast organic materials might be destroyed at the surface. 68

* Sol is used to designate the Martian day, which is 39.6 minutes longer than an Earth day; 20 July was listed as sol O because a few hours were left in the sol (local lander time) at the time of landing. Sol 1 began late on 20 July, at the first lander 1 midnight.
** The relay links for the first 11 sols were pre-programmed for redundant playback and transmission to Earth of the lander-recorded data so as to prevent loss of any important information.