Last Saturday I visited Waitby Greenriggs, a Cumbria Wildlife Trust reserve where all three species are said to occur.. Its an old railway cutting with a ditch at the bottom, with Marsh helleborines in most areas, including part way up the embankments.
Fragrant orchids were everywhere. New Flora of the British Isles (3rd edn) - Stace, gives a number of parameters to distinguish between them:
|Sepals||Lab W mm||Lab L mm||Sepal W
|Sepal L mm||Spur mm
|Gc||Point Down||5.5 to 6.5||5 to 6||1||5 to 6||12 to 14||1.1|
|Gd||Horiz||6.5 to 7||3.5 to 4||1||6 to 7||14 to 16||1.8|
|Gb||Point Down||3.5 to 4||4 to 4.5||2||4 to 5||11 to 14||0.9|
Frankly there do not seem to be any other keys so this is a s good as it gets..
At Waitby I measured 16 plants, chosen pretty much at random as i walked across the reserve, though inevitably most were in the damper areas near the helleborines, though a few were on a dry bank. I measured the all the above parameters, but in addition I measured the width and depth of the central lobe (in heath fragrant the labellum is obscurely lobed), and whether there were stomata on the upper surface of the leaf (Poland and Clements suggest that only Chalk fragrant has stomata).
I then tried to sort them out, firstly by making an assessment by eye, but then using the measured parameters, I made a few adjustments to my initial conclusion. A fellow orchid enthusiast also assessed some by eye from a Facebook posting (we agreed on most but differed on a couple. What did seem to discriminate was the number of flowers on the spike (which I did not correct for height), and whilst the labellum width to length ratio was useful the separation was not as great as suggested in the literature:
|Flowers||Lab W/L||Sepal L/W||N|
Nevertheless, these three measures taken together with a ot of weight on the flower density, and the presence or absence of stomata on the upper surface of the leaf should provide sufficient to properly attribute a plant to one of the three species.
I added the number of flowers in as an afterthought, having attributed plants on all the other measures, but it looks like the most reliable single measure, particularly if corrected for height, so G. densiflora does seem very appropriate.
Of course a big underlying assumption is that I correctly identified the plants in the first place and that Waitby greenriggs does indeed have all three species. Sadly I did not have my portable DNA tool with me on this occasion to confirm my results.