St Croix Home

More about St Croix

Current Rams

Sheep for Sale

Grassfed Lamb

Breeding Issues (Conservation Genetics, Scrapie, et)

Sheep Care

Contact Us

Flock Improvement using Selection.  Part 2:  Calculations involving weights – by Karen Gerhart, Howling Oak Ranch St Croix sheep

The economic success of a commercial sheep operation is almost completely determined by just three factors:  the cost of feed and care, the total weight of lambs raised per ewe, and the price the shepherd can get for those lambs. The St Croix sheep has the advantage that it is prolific and produces vigorous lambs, increasing the number of lambs raised per ewe; however, the St Croix also has a relatively slow growth rate.  Selecting for faster growth in lambs would seem likely to improve a flock’s economic return by causing the flock to produce more pounds of lamb.  In practice, however, selection for fast growth often inadvertently favors ewes that raise fewer lambs. A shepherd who wishes to improve growth while maintaining prolificacy, fertility, and early reproduction would seem to need detailed records and a complicated selection protocol to be sure that gains in one trait were not offset by losses in another.

Selection for litter weight weaned (that is, total weight of lambs weaned per ewe) allows for substantial improvement in flock productivity without any apparent genetic losses in other traits, according to research on responses to selection using multiple sheep breeds (Ercanbrack and Knight, 1998, J Anim Sci 76:1311-1325).  [A small reduction in wool production was found but is not relevant to those of us raising hair sheep.]  Improvements in litter weight weaned correlated with improved prolificacy, greater lamb survival, higher lamb weights at weaning, higher fertility, and better survival of the ewe during pregnancy.  Total weight of lambs weaned per ewe is a very suitable trait for selection within a St Croix flock, as it is expected to allow the shepherd to use variation in one trait to maintain the strengths of the St Croix ewe (fertility, prolificacy, early reproduction and lamb vigor) while also improving lamb growth. Its calculation is also relatively simple, requiring only normal flock records plus birth weights and weaning weights for all lambs.

In a selection program, progress will be proportional to the accuracy with which we measure genetic potential for the chosen traits.  If we wish to retain the ewes that raise the most pounds of lamb, or if we wish our flock to produce faster-growing, heavier lambs, our progress will be proportional to the accuracy with which we measure genetic potential for lamb growth.  While genetics certainly contribute to the growth rate of a lamb, many environmental factors are also involved.  To the extent that we can correct for this non-genetic variation, we can improve our estimates of a sheep’s breeding value, and thus improve our ability to retain the best genetics in our flocks.  Some of the environmental factors that affect lamb growth are the age of its mother, the type of birth and rearing (single, twin, triplet, etc), and the sex of the lamb.  When comparing growth rates of lambs, it is therefore helpful to mathematically “remove” these environmental affects.  The Sheep Production Handbook has correction factors that can be used to adjust for each of these variables.  (I find it easiest to use a computer spreadsheet program like Excel to do these calculations.)

All the lambs included in a group for calculation purposes (a ‘cohort’) should be raised in a similar environment.  That is, the corrections below won’t let us accurately compare fall lambs with spring lambs, creep-fed with non-creep fed, or our flock’s lambs with someone else’s lambs.  To compare lambs within a cohort, we first need to adjust all the lambs to the same age.  Flocks that provide creep feed and wean early often wean at 90 days, so we’ll calculate the 90 day weight of a lamb, but the same calculations could be used for lambs weaned at 120 days.  First we’ll correct the recorded weights to a common lamb age.

Age adjusted weight =[(actual weight – birth weight) / (age when weighed)]* adjustment age (90 days, in this example) + birth weight

For example, to calculate the age adjusted weight for a lamb that weighed 7.7 pounds at birth, and 52 pounds at 95 days of age: [(52-7.7)/95] [ this is her growth rate] * 90 days + 7.7 pounds = 49.7 pounds.

Now that we’ve corrected for age, we can correct for environmental affects – that is, the age of the ewe, the number of other lambs she raised simultaneously, and the sex of the lamb.  The Sheep Production Handbook (1997) provides the correction factors shown in Table 1 (effectively adjusting any lamb to a single ewe lamb born to a 3-6 year old ewe).

So, if our ewe lamb in the example above was born and raised as a twin from a 3-year old ewe, we’d look up in Table 1 Lamb sex = ewe, # lambs born, raised = 2-2, and ewe age = 3  to get an adjustment factor of 1.19.  We then multiply her 90-d adjusted weight (49.7 pounds) by 1.19 to get an adjusted weight of 59.1 pounds.  This weight is expected to be a more accurate estimate of the ewe lamb’s genetic potential than was her actual weight, as it gives her “credit” for the disadvantage of being a twin.

Table 1.  Multiplicative adjustment factors for adjusting lamb pre-weaning and weaning weights to a common age of dam, lamb sex, and lamb type of birth-rearing. Lambs born in litters of more than 3 should use the triplet adjustment factors.  Lambs born as singles and reared as twins should use the twin (2-2) adjustment, and lambs born as singles or twins and raised as triplets should use the triplet (3-3) adjustment factors.

Total Weight of Lambs Weaned per Ewe

Once the lamb weights have been adjusted for age and maternal affects, add all the lamb weights for each ewe together to get total pounds of lamb raised per ewe.  An example of some typical results from our flock is shown below (Table 2).  It appears that ewes C and D are performing well, while the performance of ewes A and B is less impressive. While somewhat informative, one season’s weights have limited value as they provide an extremely narrow view of a ewe’s production. 

Table 2.  Total weight of lambs weaned per ewe for one season (data from St Croix ewes at Howling Oak Ranch).

We pool data across several breeding seasons for each ewe in an effort to increase the reliability of our estimates of ewe productivity (Table 3).  To do so, we must treat different cohorts of lambs as roughly comparable, and if our management or the food resources available to the ewes and lambs vary dramatically from season to season, this may not be a legitimate assumption.  That risk is offset somewhat if all the ewes compared have multiple seasons of data collected, so that each ewe has likely been sampled under better and worse conditions.  Certainly multiple samples per ewe allow us to draw much firmer conclusions about an individual ewe’s performance.  In this instance, I would conclude that Ewe A is young and promising; Ewe B is a solid

Table 3.  Total weight of lambs weaned per ewe over several breeding seasons, along with the across-season average for each of 4 ewes (data from St Croix ewes at Howling Oak Ranch).

producer; Ewe C is exemplary; and Ewe D is lagging in production.  Therefore, daughters of ewes A, B and C are good candidates for replacement ewes, while sons of ewe C are good candidates for flock sires.  Ewe D is a candidate for culling and her sons and daughters are also unlikely to perform well.

Total weight of lambs weaned per ewe is my preferred selection criterion because this trait encompasses several reproductive characteristics important to the St Croix breed, including ewe fertility and prolificacy, lamb survival, and lamb growth.  Selection based on the single trait of individual lamb weaning weight will likely result in more rapid improvement of lamb weaning weights within a flock, but may also cause erosion of other important reproductive traits such as prolificacy.  Nonetheless, it too is easily calculated and often used in selection programs.

Weights of individual lambs

Lamb weights can be compared at about 60 days of age, when they are an index of a ewe’s milk production, or lamb weights can be compared at or after weaning.  The later weights more accurately reflect the lamb’s genetic potential for growth, in that they are less influenced by the performance of the ewe.  To compare growth among lambs, first calculate adjusted lamb weights as in the previous example.  Then calculate the average lamb weight, and subtract this average from each lamb’s weight to get the number of pounds each lamb outperformed (or underperformed) compared to average:

Weight difference between the ewe lamb (from the beginning of this article) and the average lamb: adjusted weight of ewe lamb = 59.1 lbs; average adjusted weight of all lambs in the cohort =  54.0 lb; 59.1 – 54.0 = 5.1 pounds.

We can calculate a rough Estimated Breeding Value (EBV) by multiplying the weight calculated above by the heritability for this trait.  An EBV gives us an idea of this ewe’s genetic merit.  The heritability for 90-day weights is 25%, or .25 (from Table 4).

EBV = (difference between individual and cohort average) * heritability of trait = 5.1 pounds x .25 = 1.3 pounds. 

We are almost done.  However, this ewe will contribute only one-half of the genetic influence on her progeny (with the ram contributing the other half), so we can convert the EBV into an EPD (estimated progeny difference) by dividing by 2. 

EPD = EBV/2 = 1.3 pounds / 2 = 0.65 pounds.

Thus, this ewe lamb is expected to produce lambs that weigh about 0.6 pounds more than the average lamb in our flock at 90 days of age.

This EPD is similar to those calculated by the National Sheep Improvement Program (NSIP) – except that the NSIP program improves the accuracy of the EPD estimate by using complex mathematics involving information on the ewe’s relatives and related maternal traits.

A simpler alternative to calculating an EPD is to divide each lamb’s weight by the average lamb weight, to get a percent performance; lambs with a score of above 1 demonstrated above average growth.  For our sample ewe lamb:  59.1/54 = 1.09 so this lamb is 9% larger than the average lamb (or 109% of the average lamb’s weight).

Table 4.  Heritabilities of Various Traits (from Sheep Production Handbook, 1997, p. 61).  Higher values mean that the trait is more heritable, while lower values mean that environmental affects tend to overwhelm the affects of genetics in determining the phenotype.

Summary Various sorts of information collected during routine record keeping, as well as lamb weights, can be used to help a shepherd determine which ewes in the flock are most productive and which should be culled.  Selection will be most rapid when the data are corrected for environmental influences and reflect the genetics of the individual accurately.  Single-trait selection can result in rapid progress, but includes the risk that other important traits are being lost.  Total weight of lambs weaned per ewe encompasses various aspects of reproduction; selection based on this trait is suitable for the St Croix breed, as it is expected to improve individual lamb weaning weights without sacrificing other important aspects of lamb production.  Rough EPDs can be calculated using lamb weights and heritability estimates.

Printable version