“No, that’s preposterous,” broke in Barnett. “If we’re to take it that there’s been no slip-up in the photography, then surely there’s only one possible explanation. The cloud is moving towards us. In the second slide it’s nearer to us, and therefore it’s obscuring more of the distant stars. At what interval apart were the two plates taken?”

“Rather less than a month.”

“Then there must be something wrong with the photography.”

“That’s exactly the way I reasoned last night. But as I couldn’t see anything wrong with the plates, the obvious thing was to take some new pictures. If a month made all that difference between Jensen’s first plate and his second, then the effect should have been easily detectable in a week — Jensen’s last plate was taken on 7 January. Yesterday was 14 January. So I rushed up to Mount Wilson, bullied Harvey off the 60-inch, and spent the night photographing the edges of the cloud. I’ve got a whole collection of new slides here. They’re not of course on the same scale as Jensen’s plates, but you’Il be able to see pretty well what’s happening. Put them through one by one, Bert, and keep referring back to Jensen’s plate of 7 January.”

There was almost dead silence for the next quarter of an hour, as the star fields on the edge of the cloud were carefully compared by the assembled astronomers. At the end Barnett said:

“I give up. As far as I’m concerned there isn’t a shadow of a doubt but that this cloud is travelling towards us.”

And it was clear that he had expressed the conviction of the meeting. The stars at the edge of the cloud were being steadily blacked out as it advanced towards the solar system.

“Actually there’s no doubt at all about it,” went on Marlowe. “When I discussed things with Dr Herrick earlier this morning he pointed out that we have a photograph taken twenty years ago of this part of the sky.”