Featured Image: Lick Observatory Records: Photographs. UA36 Ser.7. Special Collections and Archives, University Library, University of California, Santa Cruz.

The British expedition to Sobral in Brazil and to the island of Principe in 1919 gave the first evidence in favour of Einstein’s theory of general relativity. However, many scientists wanted stronger evidence before they would be convinced of the theory’s validity. The British expedition’s analysis relied on only a small number of stars, and on the ground, the two expedition parties encountered many difficulties, which led to large errors bounds in the results. This, whilst making it clear that gravity leads to a deflection of light, and that it likely was not to the degree predicted by Newtonian mechanics, was unclear on whether the precise value of deflection agreed with the theory of relativity. Scientists wanted a definite answer.

Campbell also had considerable influence on the day-to-day running of the Lick Observatory back in California. Being on the summit of Mount Hamilton, the observatory was secluded from the rest of the world and practically required its workers to live there – establishing an isolated scientific community. To be present at the 1898 expedition to India, she had left her two young children at home (of course in the hands of competent carers) for 8 months, a reflection of the fondness of her marriage to William Campbell. Her influence on the work of the observatory was very much unheard of in most professional scientific institutions. Elizabeth Campbell’s work was generally not scientific, as she had taken one astronomy module at university but graduated with a degree in English. In 1922 however, she would be essential to sustaining the team on the site, and would help with both operating the spectrograph and developing the photographic plates taken during the eclipse.

Wallal is an extremely remote location, next to the Eighty Mile Beach and west of the Great Sandy Desert. It is 110,000 square miles in size and one of the hottest places in Australia. This location had favourable weather forecasts for the eclipse, although it was extremely difficult to access. On request by Father Pigot from the Riverview College Observatory in Sydney, the Commonwealth Government agreed to provide a vessel from the Australian Navy to transport the observers, their equipment and supplies to and from the port of Perth. News of this service was promoted across the world to encourage teams to take advantage of the opportunity. At the site, the Nyangumarta people assisted the teams by carrying their equipment inland and onto donkey wagons.

This courtesy from the government allowed the plans for the eclipse to be scaled up dramatically. The Lick Observatory enlarged the scale of their expedition, a decision which they did not regret. An expedition from the Kodaikanal Solar Observatory Observatory in southern India, which had initially planned to travel to the Maldives, adjusted their plans and instead joined the parties in Wallal. A privately organised British expedition also joined.

Other expeditions were also dotted across the line of totality. The Observatory in Adelaide, southern Australia sent a party to the Cordillo Downs, located 1500 miles away. Transporting their supplies and equipment would take over a month to do by camel train. The Sydney Observatory sent a team to Goondiwindi in Queensland, located in eastern Australia, in an expedition organised and financed by Sir Wilfrid Russell Grimwade.

Travelling to Christmas Island were expeditions led by the Greenwich Observatory, England and a joint Dutch-German party. The Greenwich party were too far in their preparations to take up the generous offer by the Commonwealth Government, so had to follow through with their original plan. For example, they would have had to adapt their astrographic telescope, to account for changes in latitude of the sun in different locations.

The Dutch-German expedition was led by Erwin Freundlich from Potsdam University, but the team was made up of members from multiple observatories, the navy, a mechanic and two auxiliaries. Both teams’ goals were to measure the deflection of light with greater precision than in the eclipse of 1919 by the British group. Einstein wanted to join the expedition, but he eventually decided not to go, on account of the enormous time commitment it would involve. The groups had left for the island several months in advance of the eclipse. Tragically, heavy rain set in on the day and no photographs of the sun were obtained by the Greenwich expedition. The Dutch-German expedition was able to capture some photos. The parties were devastated. Harold Spencer-Jones, leader of the Greenwich expedition, wrote that:

“Our disappointment it is impossible to describe. It seemed cruel to be sitting unable to do anything whilst the few valuable minutes for the coming of which so much preparation had been made were passing beyond recall.”

Jones, J.S. (2022) ‘The 1922 Solar Eclipse at Christmas Island: “our disappointment it is impossible to describe”’, Journal of Astronomical History and Heritage, 25(03), p. 476. doi:10.3724/sp.j.1440-2807.2022.03.04.

The Wallal parties however were fortuned with perfect weather conditions on the day, and could therefore execute their plans perfectly. The Lick team had four cameras to perform the experiment on the deflection of light, another camera to capture photographs of the solar corona, another camera to capture the solar corona’s form, and also several spectrographs to capture the spectrum of the corona. Every expedition team that had ventured out had the deflection of stellar rays experiment as part of their scientific agenda for the eclipse. Whilst on the island, the expeditions had access to a weekly mail service delivered by air and they received wireless time signals.

In order to perform the experiment, comparison photographs needed to be captured of the star field either before or after the eclipse. To prevent systematic error, these comparisons would be taken both before and after. Since the provided transport wouldn’t operate until late August, the observers couldn’t occupy the observation site well in advance to capture the comparison photographs. They were instead taken at Tahiti, in May and June that year. The choice of Tahiti was a sensible choice to avoid introducing more error since the latitudes of Tahiti and Wallal are similar and both sites are at sea level.

Differences in atmospheric refraction of light also was an insignificant factor, which could be minimised in the later analysis of the photographs. To be confident that this wasn’t an issue, photographs of a different star field at the same latitude as the eclipse field were taken at both sites and compared. They could also detect errors since all theoretical assumptions were that gravitational displacement should be radially outward, so displacement in other directions would be clearly incorrect and could be accounted for.

The final results were released in June 1923 following a preliminary announcement in April. To analyse the results, both William Campbell and Robert Trumpler independently performed measurements on three pairs of photographic plates, and the last pair of plates were measured by Trumpler. The final reported results gave a deflection of 1.75 +/- 0.09 arcsec (a probable error of only 5%). This almost exactly matches the prediction under the theory of general relativity. The Lick team also considered whether the deflection of stars near the Sun’s poles differs from that near the equator. They found that it didn’t have an impact. Their check field photographs indicated that no instrumental errors could have given rise to the deflections that they observed and were confident in their findings. They had also paid consideration to a theory known as ‘Courvoisier’s Yearly Refraction effect’, which suggested that deflection could be explained independently of general relativity theory. It predicted a dramatically different deflection of light and of course, was dismissed by many scientists once the results from the Lick team were released. The scientific community finally began to truly place its faith in Einstein’s theory.