Академический Документы
Профессиональный Документы
Культура Документы
Huacaya Alpacas
by Angus McColl, Yocom-McColl Testing Laboratories, Inc., Chris Lupton,
Texas A&M University System, and Bob Stobart, University of Wyoming
186/Alpacas Magazine
Fiber Testing Terminology
Summer 2004/187
Yocom-McColl Testing Laboratories. to it. The difference in color predetermined number or type of sam-
The following information was between the standard and any ples (e.g., SE and WC), extra samples
requested with each sample: breeder given sample was expressed in were accepted from other regions (e.g.,
name; farm/ranch name, address, and terms of ∆E in CIELAB color dif- C and RM). In all, 81 samples were
contact information; animal name; ARI ference units. For white samples, tested from the Central region, 140
registration number; sex; date of birth Yellowness Indexes (YIE) and from the Great Lakes, 94 from the
or age; body weight; color (from ARI brightness were measured. Northeast, 103 from the Rocky Moun-
chart); sampling date; and, date last Discussion of the statistical model that tains, 18 from the Southeast, and 167
shorn. In this manner, 606 huacaya was used to analyze the data generated from the West Coast regions. More
alpaca samples (representing approxi- in this study is beyond the scope of female (338) than male (209 intact
mately 1.4% of animals registered in this article. Similarly, detailed discus- males, plus 56 from geldings) samples
the U.S. [2/1/2003]) were obtained sion of all the significant and non-sig- were received for testing. Fewer sam-
from 44 U.S. breeders (representing nificant interactions that were observed ples were received in the female and
approximately 1% of U.S. registered will not be discussed here. Rather, we male yearling age categories than had
breeders). The following objective plan to restrict discussion to the means been requested. In contrast, many
measurements were then conducted and distributions of the characteristics more samples were received from
on each sample using international and measured, correlations among selected mature females than had been
national standard methods. measured traits, the effects of sex, age, requested. The distribution of animal
1.Relaxed mean staple length, (MSL, location, and color on the measured ages are summarized in Figure 1 (the
standard deviation [SD], and coef- fiber properties, and finally, the overall “0” column indicates animals of
ficient of variation [CV]) was significance of our study. unknown age).
determined using 10 staples and In all, 606 samples were received and
the Agritest Staple Length Meter. Results tested. When requesting samples, we
2.Mean staple strength (MSS, SD, Since breeders from some regions were did not attempt to establish quotas for
and CV) and position of break unable (or unwilling) to submit the different colors. Natural white proved
(POB) was determined using 10
staples and an Agritest Staple
Breaker Model 2. Key to Abbreviations Used in Tables
3.Lab scoured yield was determined
using 2 X 25g subsamples, when Age, yr Age, years
this much sample was submitted. BW, lb Body weight, pounds
4.Resistance to compression was AFD, µm Average fiber diameter, microns
determined on duplicate scoured, AC, deg/mm Average curvature, degrees/millimeter
carded, and conditioned subsam- LSY, % Lab scoured yield, %
ples using an Agritest Resistance to MSL, in Mean staple length, inches
Compression Instrument. MSS, N/ktex Mean staple strength, Newtons/kilotex
5.Scoured samples were minicored N Number of samples tested
and 2 mm snippets were evaluated RTC, kPa Resistance to compression, kilopascals
for average fiber diameter (AFD, MED, % Medullated fibers (white and light fawn only), %
SD, and CV), prickle factor (PF), OBJ, % Objectionable fibers (white and light fawn only), %
comfort factor (CF), medullation
(medullated [MED] and objec-
tionable [OBJ] fibers), and average
fiber curvature (AFC and SD) Table 1. Summary data for properties measured on U.S. huacaya alpaca
using an Optical Fibre Diameter
Property N Mean SD CV, % Min Max
Analyser 100.
Age, yr 586 4.1 2.8 68.0 1 17
6.A scoured and carded (i.e.,
BW, lb 498 143.7 27.6 19.2 50.0 229.9
homogenized and cleaned) sub-
AFD, µm 606 27.9 5.3 19.2 15.1 49.3
sample was measured for color
AC, deg/mm 606 33.2 7.0 21.1 15.4 52.5
using a Spectrogard Color Control
LSY, % 605 89.8 4.5 5.0 58.3 95.0
System. For each of the three MSL, in 604 4.6 1.6 34.3 2.1 10.9
major color groups (in this part of MSS, N/ktex 605 50.4 21.3 42.4 4.9 137.8
the study: grey, brown, and black), RTC, kPa 595 5.4 0.9 15.7 2.0 7.8
one sample was chosen as the “ref- MED, % 277 17.6 11.0 62.2 0.6 61.7
erence” color and all other samples OBJ, % 277 3.8 3.7 97.0 0.1 22.0
within the group were compared
188/Alpacas Magazine
Figure 1. Distribution of animal age. to be the largest category (193 sam-
180 ples) followed by natural fawn (106),
natural brown (99), natural light fawn
160
(85), natural black (57), natural silver
140 grey (36), and natural rose grey (30).
Table 1 summarizes some of the data
120
submitted, measured, or calculated in
Frequency
Summer 2004/189
decrease. A change of 1ìm in AFD out- absolutely necessary in order to pro- on a clean basis) for excessively greasy
weighs a change in staple length of 10 vide protection against the elements. fiber. Lab scoured yields of the side
mm so far as spinning performance (of However, it is important not to select samples tested in this study were con-
wool) is concerned. for excessive grease since, all things sistently very high with the majority of
For both worsted and woolen spin- being equal, a buyer will pay less (even samples yielding over 90% (Figure 5).
ning, AFD and CV of fiber diameter
are by far the most important fiber
properties influencing spinning per-
Table 2. Correlation coefficients between selected animal and fiber traits
formance, yarn, and fabric properties.
Yarn hairiness, thickness, short-term BW AFD AC LSY MSL MSS RTC
irregularity, abrasion resistance, and Age, yr 0.37** 0.42** -0.29** 0.01 -0.39** 0.26** -0.01
yarn stiffness all increase with increas- BW, lb 0.53** -0.37** 0.19** -0.39** 0.44** 0.21**
ing AFD. As AFD increases, pilling AFD, µm -0.85** 0.24** -0.28** 0.39** 0.06
resistance improves but tear and burst- AC, deg/mm -0.24** 0.11** -0.26** 0.10*
ing strength decrease and fabric handle LSY, % 0.02 0.09* -0.00
becomes harsher. For all these reasons, MSL, in -0.55** -0.10*
and some others, premiums are invari- MSS, N/ktex 0.08
ably paid for finer animal fibers whether
it is alpaca, wool, mohair, or cashmere. *Significant correlation, P < 0.05
The range in AFD (15.1 to 49.3 **Highly significant correlation, P < 0.01
microns, Table 1 and Figure 3)
observed in this study was somewhat
larger than expected. Table 2 indicates Table 3. Tristimulus values, brightness, and yellowness indices for
that AFD is positively correlated with white alpaca samples (N = 241)
age, body weight, clean yield, and sta-
ple strength and negatively correlated Item Mean SD CV Minimum Maximum
with average cur vature and staple X (measure of “redness”) 60.71 5.63 9.27 36.15 70.35
length. Y (measure of “greenness”
and “brightness”) 62.53 6.10 9.75 35.61 73.05
Fiber curvature Z (measure of “blueness”) 58.90 7.04 11.95 28.84 71.05
Average fiber curvature is related to YIE (measure of “yellowness”) 19.15 3.02 15.76 12.74 34.62
crimp frequency (fiber cur vature (Y-Z) (measure of “yellowness”) 3.63 1.68 46.21 -0.79 12.21
increases as crimp frequency increases)
and crimp is directly related to resist-
ance to compression, a good indicator
of yarn and fabric bulk or loftiness. The Table 4. CIE Tristimulus values and color differences (∆E) for colored alpaca
average level of fiber curvature in alpaca
is quite low (compared to fine wool or Item Mean SD CV Minimum Maximum
cashmere, for example) and the range Grey (N = 50)
in values (Figure 4) is relatively narrow. L* (measure of “greenness”
Table 2 shows that average fiber cur- and brightness) 15.96 16.21 101.54 -23.87 56.54
vature is negatively correlated with age, a* (related to (X-Y)) 1.26 2.70 214.76 -1.21 9.69
body weight, fiber diameter, clean yield, b* (related to (Y-Z)) 3.72 3.86 103.89 -0.36 20.88
∆E (color difference) 20.12 12.14 60.33 0.44 57.47
and staple strength and positively cor-
Brown (N = 237)
related with staple length and resist-
L* (measure of “greenness”
ance to compression.
and brightness) -12.12 24.07 -198.53 -71.70 64.52
a* (related to (X-Y)) 4.72 6.28 133.05 -7.66 27.50
Clean yield
b* (related to (Y-Z)) 3.78 10.50 277.58 -17.65 35.30
Raw alpaca contains moisture, and
∆E (color difference) 26.71 14.30 53.52 1.20 72.84
small quantities of grease, sweat, and
Black (N = 74)
dirt, in addition to some vegetable
L* (measure of “greenness”
material. The buyer is concerned with and brightness) -4.49 14.14 -315.28 -30.89 42.19
how much clean alpaca he is buying. a* (related to (X-Y)) 1.09 2.59 237.85 -2.18 11.86
A measurement of clean yield allows b* (related to (Y-Z)) 1.30 2.48 190.49 -1.56 11.96
him to calculate that quantity. A small ∆E (color difference) 11.72 9.84 83.94 0.30 42.26
amount of grease in the fleece is
190/Alpacas Magazine
Figure 4. Distribution of average fiber curvatures. A small word of caution here may be
200 warranted. The LSY does not include
a correction for vegetable matter con-
180 tent so these yields are slightly inflated.
160 However, for most of the samples, veg-
140 etable matter was only present in very
Frequency
small quantities.
120
100 Staple length
80 Staple length and mean fiber length
60 after carding and in top are likely to
be highly correlated for “sound” alpaca
40
(MSS > 30 N/ktex). Mean fiber length
20 has a great influence on spinning per-
0 formance, yarn strength, and unifor-
20 25 30 35 40 45 50 55 mity. However, mean fiber length after
Average fiber curvature, deg/mm carding is also influenced, to varying
degrees, by crimp, staple strength, posi-
tion of break, AFD, and the degree of
fiber entanglement after scouring. All
Figure 5. Distribution of lab scoured yields. things being equal, longer fibers pro-
300 duce stronger, more uniform, leaner
yarns that have greater resistance to
250 abrasion. Once in fabric, the effects of
fiber length diminish compared to
200 effects during the spinning stage. As
Frequency
Staple strength
Figure 6. Distribution of mean staple lengths. From a processing point of view, a
300 mean staple strength greater than 30
Newtons per kilotex (N/ktex) has been
250 shown to be adequate for satisfactory
processing of wool on today’s high-
Frequency
Summer 2004/191
Resistance to compression assessed to be white, and 361 colored. have the complete range of brightness
The resistance to compression of alpaca The number of samples designated values. A range of values of (Y-Z) for
is low (Figure 8). In the commercial white for this part of the study was scoured and carded U.S. white wool
sector, RTC values > 11 kilopascals greater than the earlier number assessed has been reported to be 3 to 10. By
(kPa) are considered high, 8 to 11 kPa as “Natural White” (193) because the contrast, scoured New Zealand and
medium, and <8 kPa low. Thus alpaca assessments were made by different Australian wools are reported to fall in
is not suited to end-uses that require people. No doubt, some Natural Light the range 0 to 4. The average (Y-Z)
high resistance to compression (or high Fawn samples were included in the for all white alpaca samples is 3.63,
bulk). To produce alpaca having higher white category for color measurement. demonstrating that on average they are
RTC, it would be necessary to select for Table 3 summarizes the tristimulus as white as some of the whitest wool in
more crimp. values, brightness, and yellowness the world. Figure 9 shows that 33.7%
indices for the 241 white samples. In have values >4. YIE values for scoured
Medullation the CIE system of color measurement, and carded wools range from 20 to 28,
Most alpaca samples were medullated, X is a measure of redness, Y is a meas- so again the reported mean value for
to varying degrees. Degree of medulla- ure of greenness (also brightness), and these alpaca samples of 19.15 indicates
tion has a great influence on appear- Z is a measure of blueness. It has been they appear exceptionally white.
ance and dyeability. White and light demonstrated that for white samples, The main conclusion to be drawn
colored fibers having a medulla diame- (Y-Z) is a good indicator of yellow- from Table 3 and Figure 9 is that wide
ter greater than 60% of the width of the ness. The typical range in Y for white ranges in yellowness and brightness
fiber are chalky in appearance and also samples is 70 (very bright) to 40 (very exist for nominally white alpaca sam-
appear not to accept dyestuff. These are dull). Thus, in this set of samples we ples. Instrumentation could be very
referred to as “objectionable.” Lower
levels of medullation do not affect
appearance, though they affect other Table 5. Main effects of sex
fiber properties (e.g., strength). In this
study, medullation was estimated from
Female (N) Male (N) Geldings (N)
a measure of opacity. Consequently,
BW, lb 143.3 (287) 139.9 (160) 158.5 (50)
only natural white and natural light fawn AFD, µm 27.7b (338) 27.0b (209) 32.3a (56)
samples were reported. Compared to AC, deg/mm 33.5a (338) 33.7a (209) 28.6b (56)
apparel wool and mohair, the levels of LSY, % 89.5 (337) 90.1 (209) 90.3 (56)
medullated fiber (Table 1) are gener- MSL, in 4.3 (337) 5.1 (209) 4.3 (56)
ally high and extremely variable. MSS, N/ktex 49.6 (337) 49.9 (209) 56.7 (56)
Of course, this is normal and expected RTC, kPa 5.42 (336) 5.39 (204) 5.56 (52)
for alpaca fiber. However, the mini- MED, %* 17.7 (141) 17.6 (115) 18.1 (19)
mums for medullated and objection- OBJ, %* 3.8 (141) 3.7 (115) 4.6 (19)
able (0.6 and 0.1%) are extremely low *NW and NLF only.
and suggest that progress might be pos- a,b
Means in a row having different superscripts are different (P < 0.05).
sible if selection to reduce medullated
fiber was attempted.
Color measurement
Numerous subjectively-assessed colors Table 6. Main effects of age
are recognized by the alpaca trade.
Older than
Objective measurements of color (using
Yearling (N) 2 years old (N) 2 years (N)
a colorimeter) in terms of X, Y, Z, L*,
BW, lb 105.8c (46) 130.6b (134) 154.8a (314)
a*, and b* CIE tristimulus values will
AFD, µm 23.2 c
(59) 24.9b
(155) 29.8a (372)
(potentially) permit fleeces to be accu-
AC, deg/mm 36.2 a
(59) 35.8a
(155) 31.6b (372)
rately described in terms of this trait
LSY, % 88.5 (59) 89.5 (155) 90.1 (371)
and also permit ranges to be estab-
MSL, in 5.3a (59) 5.6a (155) 4.0b (371)
lished for specific color descriptions.
MSS, N/ktex 28.7c (59) 41.9b (155) 56.9a (371)
Color measurement of scoured “white”
RTC, kPa 5.32 (59) 5.41 (154) 5.44 (363)
wool is now an important component MED, %* 11.6b (26) 13.7b (79) 20.2a (163)
of wool marketing in New Zealand and OBJ, %* 1.9c 2.8bc 4.4ab
Australia. Six hundred and two sam- *NW and NLF only.
ples were measured using a colorime- a,b,c
Means in a row having different superscripts are different (P < 0.05).
ter. Of these, 241 were subjectively
192/Alpacas Magazine
Figure 7. Distribution of mean staple strengths. useful for establishing white lines of
120 alpaca fiber based on measures of
brightness and yellowness.
Table 4 summarizes the CIE tristim-
100
ulus data and color differences for nom-
inally grey, brown, and black samples.
80 In this method of color difference
Frequency
250
groups. Instrument measurements of
200 individual fleeces could become a very
effective tool for establishing uniform
150
color lines of alpaca if such were ever
100 deemed desirable.
Summer 2004/193
importance of average fiber diameter, Table 7 summarizes differences in weight and produce finer, longer,
some extra consideration will be given attributable to color. One indication but weaker fibers with higher fiber
to this trait later in the report. from this table is that black samples curvature. Inexplicably, the higher
Returning to Table 5, sex appears were on average coarser than white fiber curvature did not appear to pro-
not to have an effect on most of the samples with most of the other colors duce fibers having higher resistance
measured traits. Gelded males have being intermediate. Again caution is to compression.
coarser fiber than intact males and required. Before this comparison can be Table 9 shows the effects of sex and
females but again, it must be pointed properly made, it is necessary to take age on fiber diameter. For a particular
out that most of the geldings meas- into account the different ages and age group, AFD does not differ
ured were three years of age or older. sexes of the animals involved, as we do between females and intact males. In
Most of the trends observed in Table in a later analysis. Small differences fact, fiber samples from yearling and
6 appear to be intuitively correct. This occur in several of the other traits, but two-year-olds were not significantly
table quantifies the magnitudes of the again, the same caution is provided. different. Mature geldings were shown
effects of age. As alpacas age, their body Table 8 summarizes the differences to be coarser than mature females and
weights, fiber diameters, staple in means attributable to region which intact males (by about 3 microns).
strengths, and percentage of medul- reflect (to unknown degrees) the dif-
lated fibers all increase. However, fiber ferent environmental conditions Further discussion of correlations
curvature, and staple length decrease. encountered as well as the different Table 2 lists the Pearson correlation
Lab scoured yield and resistance to genetics that are present. The effects coefficients between selected animal
compression appear to be independ- of region are also confounded by and fiber traits. A knowledge of cor-
ent of age, the latter observation being age and sex. Generally, animals from relations between traits is necessary for
somewhat surprising. the WC region tended to be lighter a better understanding of the likely
194/Alpacas Magazine
Table 9. Least squares means longer wool. However, it does match six alpacas representing female, male,
of average fiber diameter our observations within the breed and castrated male registered animals
by sex and age of fine-wool Rambouillet sheep. This in the three age categories: one- and
negative correlation is also different two-year-old and adult, and then sent
Sex Age, yr AFD, µm N than that reported by other researchers to the commercial testing laboratory.
F 1 23.7c 31 for adult alpacas being farmed in Additionally, each sample was meas-
F 2 26.4c 77 New Zealand. However, most of the ured for average fiber diameter (and
F >2 29.9b 224 correlations between the remaining SD and CV), comfort factor, average
M 1 24.3c 28 fleece characteristics are very similar fiber curvature (and SD and CV),
M 2 25.8c 75
in magnitude and significance to medullation (white and light fawn sam-
M >2 30.8b 95
those reported in the present study. ples only), lab scoured yield, average
G 1 — 0
Coarser alpaca tends to yield higher staple length (and SD and CV), staple
G 2 24.8c 3
and be stronger. strength, position of break, resistance
G >2 33.1a 52
In the populations tested in New to compression, color differences (col-
a,b,c
Means in a column having different
Zealand and the U.S., AFD and ored samples), and brightness and yel-
superscripts are different (P < 0.05).
RTC were not significantly correlated. lowness (white samples). Compared to
This was not entirely predictable. wool of similar fineness, alpaca was
results of selective breeding. A corre- In wool, for example, finer fibers tend shown to be much higher yielding,
lation coefficient (r) of 1 between two to have more crimp that results in a more heavily medullated (a distinctive
characteristics indicates that one char- positive, significant correlation with feature of alpaca), longer, and consid-
acteristic may be used to perfectly pre- RTC. However, truly fine wools also erably stronger. Resistance to com-
dict the other (a rare occurrence in exist, having ver y bold crimp and pression was invariably lower for alpaca
nature). An r value of 0 indicates there predictably, these wools have relatively compared to wool of comparable
is no relationship between the two low RTC. The AC turns out to be a dimensions, due primarily to the gen-
characteristics. A P value of < 0.05 indi- slightly better predictor of RTC, but is erally lower levels of crimp in the alpaca
cates this is not a chance occurrence negatively correlated with LSY and fibers. Less crimp results in leaner,
and that a significant relationship exists. MSS. The remaining correlations smoother, less bulky yarns and fabrics,
Thus AGE is positively and significantly between LSY and MSL, MSS, and an attribute for worsted (but not
correlated with BW, AFD, and MSS RTC are all very low, although the rela- woolen) constructions. In addition,
and negatively and significantly corre- tionship with MSS is just significant (P data were analyzed (results to be pre-
lated with AC and MSL. Correlations = 0.03). MSL was shown to be nega- sented in a refereed journal article) in
between AGE and LSY and AGE and tively and quite highly correlated with terms of sex, age, region, color, and
RTC are not significant (P > 0.8). MSS and lowly correlated with RTC. their interactions.
Interpretation: as animals age, BW, Finally, MSS and RTC had a very small,
AFD and MSS increase, AC and MSL positive correlation. Acknowledgement
decrease and LSY and RTC are unaf- Funding by the Alpaca Research Foun-
fected. These trends seem intuitively Summary dation is gratefully acknowledged. For
correct (with the possible exception of A study was conducted to establish a complete copies of our Final Report,
RTC) and are somewhat similar to comprehensive profile of U.S. hua- please contact Patricia Craven, Presi-
trends observed for sheep and Angora caya alpaca fiber characteristics that dent, Alpaca Research Foundation
goats, as are many of the correlations will be useful for educational, pro- at alpacone@microserve.net.
listed in Table 2. motional, policy, selection, and breed-
The AFD, LSY, MSS, and RTC are ing purposes. Specifically, the ranges, Angus McColl, Yocom-McColl
positively correlated with BW, while means, and distributions of all impor- Testing Laboratories, Inc.
larger BW are associated with lower AC tant fiber characteristics and body In 1963, Angus McColl and Ira Yocom
and MSL. This latter negative correla- weights of U.S. alpacas were meas- founded Denver-based Yocom-McColl
tion would not be expected for wool. ured and calculated using interna- Testing Laboratories, Inc. as an inde-
The highest reported r value is a nega- tionally accepted objective test pendent wool and animal fiber testing
tive correlation between AFD and AC, methods. Animals in specified age facility. Angus is a member of the
i.e., as fibers become coarser, curvature ranges and of known sex represent- American Society for Testing and
decreases, i.e., crimp frequency declines. ing six geographical regions in the Materials (ASTM) and the International
The AFD is also negatively corre- U.S. were weighed and sampled in Wool Testing Organisation (IWTO).
lated with MSL. This observation does approximate proportion to their pop- Yocom-McColl Laboratories utilizes
not match those made among different ulation density in the respective ASTM and IWTO procedures and meth-
breeds of sheep for which we notice regions. Fiber samples were shorn ods when testing fibers and operates
the coarser breeds generally grow from the mid-side of six hundred and Sirolan LaserScan and OFDA instru-
Summer 2004/195
ments, as well as microprojection for
the measurement of fiber diameter and
distribution. He can be reached at (303)
294-0582 or ymccoll@ymccoll.com.
196/Alpacas Magazine