It seems a highly improbable thing to do, but three months out from the event, STV have commissioned a poll asking solely about European Parliament voting intention. The results provide a degree of corroboration for the recent YouGov poll showing a Labour lead of seven points.
Liberal Democrats 10%
There are of course a number of 'buts' here - some objective, some intuitive. The most obvious objection is that STV (perhaps for reasons of cost?) chose to commission Progressive Scottish Opinion, a firm that is not a member of the British Polling Council and that has a track record of inconsistency and unreliability. In the run-up to the 2007 Holyrood election, if memory serves me right PSO polls had the SNP 12% up one week, 3% behind the next, and 6% ahead the week after. A possible explanation for this is that PSO do not weight by past vote recall, as Mike Smithson points out tonight.
The intuitive objections are that the Conservatives are simply too low and Labour are too high for this poll to be believable. Labour received just 26% of the vote at the last European elections - OK, they were relatively unpopular during the immediate aftermath of the Iraq war, but does anyone seriously believe they're 15% more popular now? And whatever the Tories' ongoing difficulties in making a meaningful breakthrough in Scotland, I think we can safely assume that their share of the vote will be going up in June, not down.
My final observations are ones I've made before. Any poll for the European elections - even one conducted by a more credible company - should be taken with a heavy dose of salt at this stage. That's partly because turnout is likely to be almost absurdly low, which always increases the level of unpredictability. And secondly, history shows the PR list system results in the larger parties doing significantly less well than predicted, while fringe parties benefit from a rare moment on a more level playing-field.
My own guess - if Labour are ahead in June, their share of the vote will be in the low thirties at the absolute most.