Natural Selection at the Edge of Life: Allelic Polymorphism and Recruitment in High Latitude Arctic Char (Salvelinus alpinus) Generated and Maintained by Environmental Extremes

Round 1

Reviewer 1 Report

The paper presents an interesting study of the effect of ambient temperature on the strength of year classes of Arctic charr from high latitudes and on the frequencies of their esterase-2 alleles. Charr were sampled in 1979-1981 in eight lakes of Svalbard and separated into year classes based on otolith age. The author found clear positive correlation between relative year class size and mean air temperature during two summers preceding the spawning and between the frequency of Est-2*100 allele and the temperature in the year of hatching. The first effect demonstrates the influence of climate on parental reproductive parameters, in particular the dynamics of maturation, spawning frequency, fecundity and the quality of gametes resulting in low recruitment following cold years and high recruitment following warm ones. The second effect evidences temperature-dependent selection acting on the isozyme loci after hatching.  Based on the observed phenomena, the author considers climatic fluctuations as a factor sustaining allelic polymorphism in Arctic charr populations. He also forwards an interesting hypothesis of genetic isolation by the year of recruitment between sympatric charr forms in one of the lakes under study. One more important observation is the tendency towards the fixation of alternative Est-2*100 and Est-2*90 alleles in anadromous and landlocked charr. The obtained results give insights into aspects of charr life history, physiology and speciation; they demonstrate that although nowadays allozyme markers are rarely used in genetic studies of fishes, they are still an important tool for the investigation of their ecology and evolution.   

The paper is informative, concise and well written; I have only a few minor comments. In the title, only one of the two studied phenomena is mentioned (allelic polymorphism). Perhaps it would be worth adding “recruitment fluctuations”?  Speaking about isolation by the year of recruitment between large and small morphs in Lake Johansvatn (lines 231-242) the author suggests that “the lower modal group with high allele frequency (0.74±0.15) was born during a series of warmer years or is the offspring of a rich year class of parents born during a series of such years, while the low frequency group (0.34±0.17) has the opposite origin.” This mechanism is not quite clear because both morphs (though not all adult individuals) obviously spawn every year and thus generations of cold and warm years are not reproductively isolated like generations of odd and even years of pink salmon. The recruitment should be low after cold years and high after warm years in both groups. This point needs a bit more detailed explanation.

Line 28: Mechanisms… is àMechanisms… are

Line 31: Under what circumstances do natural selection maintain à Under what circumstances does natural selection maintain

Line 73: Arctic char are iteroparous, and spawn in autumn -  Since some populations spawn in spring, it is better to say “and typically spawn in autumn”

Line 125: Do you mean Pearson’s correlation coefficient r or coefficient of determination R²? This is not quite clear because in the Results section (line 146 and below) you use denotation r².

Figure 2 – If vertical axis is the relative year class size, i.e. the proportion of different age classes in percent, then it should be named Relative year class size (%), not …(n)

Line 159: A) relative year class strength – it looks like in Fig. 3A not relative, but absolute year class sizes are used, please check

Table 2 – it seems that column names p  and 95% c.i. are swapped

Табл 3 – The first column should be named  “Age and sex groups” and “Age groups, total” should be moved below the column names line. Also, the last column should be obviously named 95% c.i., not p.

Line 218: may need a least another year à may need at least another year

Lines 228-229: selection processes influencing year class strength and allele frequency is proposed to grant à are proposed to grant

Line 250: Koehn [70] provided… à Koehn [30]? provided

Line 279: in a recent study of cisco… [63] – “recent” is hardly applicable here since this paper was published in 1991

Line 287: and year class allele frequency after hatching à and year class allele frequency with temperature after hatching

Lines 428-433 – numbers of the last three references should be 63, 64, 65

Author Response

Thanks for finding all those grammatical errors and your kind comments.

Speaking about isolation by the year of recruitment between large and small morphs in Lake Johansvatn (lines 231-242) the author suggests that “the lower modal group with high allele frequency (0.74±0.15) was born during a series of warmer years or is the offspring of a rich year class of parents born during a series of such years, while the low frequency group (0.34±0.17) has the opposite origin.” This mechanism is not quite clear because both morphs (though not all adult individuals) obviously spawn every year and thus generations of cold and warm years are not reproductively isolated like generations of odd and even years of pink salmon. The recruitment should be low after cold years and high after warm years in both groups. This point needs a bit more detailed explanation.

Thanks! I have removed the reference to the paper on reproductively isolated like generations of odd and even years of pink salmon, and reformulated the text.

Line 28: Mechanisms… is àMechanisms… are

Thanks! The text has been changed.

Line 31: Under what circumstances do natural selection maintain à Under what circumstances does natural selection maintain

Thanks! The text has been changed.

Line 73: Arctic char are iteroparous, and spawn in autumn -  Since some populations spawn in spring, it is better to say “and typically spawn in autumn”

Thanks! The text has been changed.

Line 125: Do you mean Pearson’s correlation coefficient r or coefficient of determination R²? This is not quite clear because in the Results section (line 146 and below) you use denotation r².

I meant R². Thanks! The text has been changed.

Figure 2 – If vertical axis is the relative year class size, i.e. the proportion of different age classes in percent, then it should be named Relative year class size (%), not …(n)

The vertical axis show numbers of fish recorded from the various cohorts.

Thanks, the figure has been changed, and replaced.

Line 159: A) relative year class strength – it looks like in Fig. 3A not relative, but absolute year class sizes are used, please check

Thanks! No the size of year class strength was not relative! The text in the legend has been changed.

Table 2 – it seems that column names p  and 95% c.i. are swapped.

Thanks. The table has been updated

Таble 3 – The first column should be named  “Age and sex groups” and “Age groups, total” should be moved below the column names line. Also, the last column should be obviously named 95% c.i., not p.

Thanks! The column title has been altered, and the table updated.

Line 218: may need a least another year à may need at least another year

Thanks! The text has been changed.

Lines 228-229: selection processes influencing year class strength and allele frequency is proposed to grant à are proposed to grant

Thanks! The text has been changed.

Line 250: Koehn [70] provided… à Koehn [30]? Provided

Thanks! The text has been changed.

Line 279: in a recent study of cisco… [63] – “recent” is hardly applicable here since this paper was published in 1991

Agree! Thanks! The text has been changed.

Line 287: and year class allele frequency after hatching à and year class allele frequency with temperature after hatching

Thanks! The text has been changed.

Lines 428-433 – numbers of the last three references should be 63, 64, 65

Thanks! The reference list has been updated, and the numbers have been changed.

Reviewer 2 Report

This article presents a rather complex investigation, carried out by using multiple techniques, supporting temperature variation to generate and maintain polymorphism in Arctic char. Considering the robustness of the investigation, the novelty of the results, the quality of the presentation, and the ongoing climatic situation, this manuscript is surely of interest of an international and potentially wide readership.

Overall, the manuscript is well written and it is complete of all needed information. Only one minor modification is suggested

Figure 1. Please, add geographical coordinates

Author Response

Thanks for your comments.

I have changed the legend to Figure 1 to “The geography of Svalbard at 74-81°North and 9-33°East , with sampling sites and weather stations”.