Photosynthesis, Respiration, and Modeled Growth Rate in the Daytime and the Night on a Basis.

**(A)** Photosynthesis (*A*, pale color) and respiration (*R*, dark color) were measured at the growth temperature in five different thermocycles. The rates are shown on a per day basis, after correcting for the length of the light period (8 h) and the night (16 h). The net diurnal C gain is the difference between *A* and *R*.

**(B)** The was estimated from the difference between biomass at harvest () and biomass at transfer to the thermocycle treatments at 21 d.

**(C)** Correlation coefficient between the net diurnal C gain and . Note that C accounts for ∼42% of the , so the numbers on the *y* axis must be multiplied by 0.42 to allow a comparison of the absolute rate of use of C for growth.

**(D)** Estimated rate of growth in the daytime and the night. The rate of growth in the day is estimated as *A* minus the sum of C accumulated in starch, sugars, organic acids, and amino acids (), divided by 0.42 (the proportion of C in ). The rate of growth at night is estimated as the sum of C accumulated in starch, sugars, organic acids, and amino acids minus *R*, divided by 0.42.

Data represents the mean ± sd (*n* = 4), with one replicate comprising five pooled rosette plants. Orange, ; green, ; blue, . For details of the calculations, see ; seeTable 1 andSupplemental Data Set 3 online. One-way was used to identify potential candidates for a statistically significant difference in *A* and *R* between treatments separately for each of the three accessions and two time points. After P value correction using Holm’s method (P < 0.05), individual contrasts were then identified in a post-hoc Tukey test (P < 0.05). They are indicated by different letters within the same time point (, lowercase; , uppercase). Error bars and significance tests are absent for **(B)** and **(D)**, where the calculations are based on average values for *A*, *R*, and summed C.

## PubMed Commons