Skip to main content
. 2009 Jan 1;105(2):291–300. doi: 10.1093/aob/mcp288

Table 2.

Best-fit stepwise regression models explaining extensions of the spore cloud (in mm), and simple models including only the linear relationship to capsule diameter as a predictor, at different post-release stages in Sphagnum (n = 27, except for hspike with n = 28)

Cloud extension variable Model Cumulative R2 of predictors (%) Simple linear model
Spike height, hspike −221·1 + 268·9 × D − 54·2 × D2 + 13·6 × sforest 62·7***; 67·0*; 72·9* −9·1 + 53·7 × D
Centroid height, hcen −69·9 + 158·3 × D − 1·8 × d + 11·5 × sdark − 29 × D2 58·2***; 70·8**; 74·4*; 77·8(*) −2·2 + 41·8 × D
Peak height, hpeak 63·5 + 70·0 × D − 2·0 × d 60·5***; 64·7(*) 8·5 + 69·0 × D
Width at tcen, wcen −43·1 + 95·6 × D − 0·9 × d − 17·0 × D2 65·3***; 69·7(*); 72·4(*) −3·6 + 28·9 × D

D, Capsule diameter (mm); sforest, swamp forest species (1/0; dummy variable); d, spore diameter (μm); sdark, dark spores (1/0; dummy variable).

***, P < 0·001; **, 0·001 < P < 0·01; *, 0·01 < P < 0·05; (*), 0·05 < P < 0·15.

Cumulative R2 is the additive explanation of each of the introduced predictors, in the order added to the model. All added predictors resulted in higher model adjusted R2.