We thank Paterson et al. (2011) for their comment on Scott and McCarthy (2010). As stated in our original paper, we agree that Lake 227 has remained eutrophic in response to phosphorus (P) fertilization alone. However, the effect of halting nitrogen (N) fertilization cannot be fully evaluated until the N pool in Lake 227 has reached steady state. The relatively high chlorophyll a (Chl a) and phytoplankton biomasses reported for 2007and 2009(Paterson et al. 2011 were not surprising considering the tremendous variability in these measurements over the entire experimental period. However, these new data do not nullify the decreases in total N (TN) and Chl a from 1990 to 2005, determined from data in Schindler et al. (2008). Additional years of P fertilization at the current rate are needed to determine if decreasing N, but stable P, results in measurable decreases in Chl a or phytoplankton biomass.TN concentrations in Lake 227 have continued to decline, but total P (TP) concentrations have not changed since N fertilization was suspended in 1990. Furthermore, dissolved inorganic N (DIN) : total dissolved P (TDP) and total dissolved N (TDN) : TDP ratios have decreased, and heterocyte abundances have increased since 1990 (Paterson et al. 2011). Paterson et al. (2011 misinterpret the increase in heterocyte abundance as a signal that the phytoplankton have overcome N deficiency. The mere presence of heterocystous cyanobacteria does not indicate that N fixation is meeting phytoplankton N demand (Lewis and Wurtsbaugh 2008). If N-fixing cyanobacteria were supplying N at a rate proportional to P fertilization, TN concentrations would remain constant or increase relative to TP. Instead, N losses from denitrification, burial, or washout have been greater than ecosystem N gains from N fixation, which confirms that fixed N has not accumulated in Lake 227 over a multi-annual timescale (Scott and McCarthy 2010).The energetic costs of N fixation limit phytoplankton primary production and biomass accumulation. Many studies have shown that cyanobacteria differentiating heterocytes and fixing N have lower growth efficiencies and accumulate less biomass per unit N than those assimilating DIN (Rhee and Lederman 1983;Attridge and Rowell 1997;De Nobel et al. 1997). Therefore, the tradeoff associated with the high energetic costs of N fixation, compared with the relatively low energetic costs of DIN assimilation, will yield less biomass when cyanobacteria are fixing N, even at the ecosystem scale.Using a meta-analysis of almost 900 studies on freshwater and marine environments, Elser et al. (2007) showed that the combination of N + P enrichment always yielded more biomass than N or P enrichment alone. The tremendous temporal variability in Chl a and phytoplankton biomasses over the 41-yr experiment in Lake 227 masks this effect. However, when data were grouped by the seven unique combinations in P and N fertilization rates (Schindler et al. 2008;Paterson et al. 2011), average annual Chl a concentrations were strongly proportional to annua...