SEM scanning demonstrates that thecuticular cell length, height, and scaleedge angle of suri alpaca fiber ismeasurably different from huacayaalpaca, cashmere, wool, and othermembers of the Camel Family.

The Washington State Disease Diag-nostic Lab (WADDL) in Pullman, WAanalyzed 35 suri alpaca fiber sampleswith surface scanning electron micro-scopy in February through November2005 for the authors. The WADDLnormally uses the Electron Microscopyand Imaging Center (EMIC) to identify virus and bacteria species forthe state of Washington. Samples were prepared by EM Supervisor ChrisDavitt, Ph.D., who took two micro-graphs of each fiber sample.

To provide a baseline to other membersof the Camel Family, 19 llama, eighthuacaya, six vicuña, five guanaco, andtwo de-haired Bactrian camel sampleswere also analyzed. Other specialtyfibers analyzed included one sampleeach of white angora rabbit, washedfawn mohair, and Bombay silk. Twosamples each of Soft Rolling Skin®

(SRS) merino wool, and both domestic(in the grease) and very fine, washedwhite Chinese cashmere were tested.

SEM scanning demonstrates that thecuticular cell scale length, frequency,height, and scale edge angle of surialpaca fiber is measurably differentfrom these other specialty fibers andother members of the Camel Family.Cuticular scale length is expressed asthe Mean Scale Frequency (MSF) per100 micron (µ) field of view as meas-ured by the SEM. High-luster suriappears to be most similar to cashmere,though it has an even longer and lowerscale height. This study has importantimplications for the alpaca industry,including AOBA and the AFCNA. Surialpaca breeders can now claim to pro-duce a natural fiber which has lusterthat is equal to or greater than cashmere.Due to its very low cuticular scaleheight, both suri and huacaya breederscan explain why their products havesuperior handle, compared to wool ofsimilar average fiber diameter (AFD).

Sample Demographics

Samples were collected from 35 surialpacas, comprised of 20 male and 15 females. Thirty of the suris were

Angstrom (Å): A unit of length, equal to one ten-millionth of a millimeter, primarily used to express electromagnetic wavelengths.Compound Microscope: An optical microscope with two ocular lenses.Cuticular Cell Scale: The external-most structure of hair. Highly Evolved: a merino or huacaya that has a high frequency of crimp, or a suri with unusually high luster or tightly-penciled locks.Mean Scale Frequency (MSF): The number of cuticle scales per 100 micronthat are the mean of the data set.Micrograph: A digital photograph of the SEM sample.Micron: 1/1,000,000 of a meter (1/10,000 of a millimeter, or about 1/25,000th ofan inch). Scanning Electron Microscopy (SEM): Use of the electron microscope toview the surface characteristics of an object.South American Camelid (SAC): A literal translation of Camelidos Sud Amer-icanos, incorporating all the members of the Lama genus, including thealpaca, llama, guanaco, and vicuña.Transmission Electron Microscopy (TEM): Use of the electron microscopeto view an object on-end, rather than on the surface of the subject.

Terminology

Surface Scanning Electron Microscopy of Suri Alpaca Fiber and Other Members of the Camel Family

By Andy and Dr. Cheryl Tillman

SCIENCE

Suri fiber is characterized by an indistinct scaleedge that is difficult to visualize even with digital imaging tools.

158 ■ Alpacas Magazine

offspring from the 1991, 1996, and1998 Bolivian importations to theUSA. These animals were randomlyselected from our own herd based onwhether we had fiber samples or fleecessaved from the previous year’s clip.Three Australian suris were also tested.These were first-generation suris from asuri x huacaya cross that were pheno-typically suri. Two more of our suriswere from Peruvian origins. Fiber sam-ples were collected in two phases. Thefirst phase of 14 samples was collectedfrom shorn virgin fleeces. The remain-ing (majority) of samples were takenfrom unshorn suris using a uniform

“Yocum-McColl” sample site used forlaser scanning. The second phase offiber samples had a slightly lower MSFthan the first. Twenty of the surialpacas had a uniform age range of 11to 13 months at the time of sampling.Eight samples from Phase 2 were sec-ond or third fleeces, rather than virginfleeces (and these animals were three tofive years old). Three of these could becompared to the same animal’s virginfleece. MSF was within +/- 10% of thevirgin fleece. No significant differencewas noted between the first and subse-quent fleeces. The remaining seven animals were 21 to 25 months old.

The narrow flat lock was highly correlated toluster in this study.

A well-locked suri in his second fleece, likethis animal, typically had an MSF of 7-7.5.

Spring 2006 ■ 159

Most of the straighter fiber samples had excep-tional luster and a low MSF.

The Long Smooth Scale™ of suri fiber canclearly be seen in this micrograph. Averagescale length was 16.25µ. The debris in back-ground is dandruff.

According to ARI statistics, over half ofall suri alpacas in the United States arecolors other than white. About two-thirdsof this study were colored: eleven subjectswere white, eight brown, seven black,five fawns, three greys, and one beige.

The majority of samples (21) had atwisted style of lock, including severalwith a highly-evolved “pearl” lock of 3-4 very small inter-twined locks. Thestyle of lock ranged from cotted to“independent.” The remaining sampleshad either a narrow flat lock (3), a rela-tively straight fleece that clung closelyto the body (6), a fan-shaped lock (3),or a coiled (corkscrew) lock (2). Fourof the donors were overweight, andthese had the strongest Average FiberDiameter (AFD). The samples werelaser scanned and had an AFD of 16.7-32.8µ with a mean of 22.8µ and10.3% coarse fibers. All of the subjectanimals were healthy, de-wormed, bodyscore condition 7-9, and compared tothe population at large, free of heatstress (we shear). None of the donorshad ever been washed with shampoo or conditioners.

Why SEM?

The SEM micrographs provide detailnot obtainable with a conventionalcompound microscope. An opticalmicroscope uses visible light of a wave-length of several thousand angstroms(Å). Such an instrument is actually aphoton microscope, since a ray of lightis a beam of photons. An electronmicroscope uses a beam of electronsinstead of a beam of light.

The main advantage of the EM micro-scope is its potential for very highresolving power. This is based on thepossibility of using electrons whose deBroglie wavelengths are less than 1Å.Objects as small as 2.3Å have beenresolved, a feat forever beyond the capa-bility of a microscope using visible light.1

Research Methodology

Micrographs were taken at an accelerat-ing voltage of 15 Kilo Volt (KV) and1,000 X. The quality of the micrographstaken by a skilled EM instructor like Dr.Davitt made it possible to accurately

measure the length of each scale on afiber, the height of scale, frequency ofscale, angle of scale, and fiber diameter.For the purposes of this article, the rel-atively simple International Wool Tex-tile Organization (IWTO) DTM-XX-97methodology was used with the excep-tion of viewing the samples at 1,000 Xmagnification rather than 600 X. Thehigher magnification was necessary foraccurately measuring suri scale height,which is almost impossible to measureeven with digital imaging tools. Woolhas a scale height of <8 micron and caneasily be measured at 600 X.Measuring scale height is important,since scale height is one reason surialpaca fiber has a low coefficient offriction and feels finer than it is.

Following IWTO DTM-XX-97methodology, the authors counted thenumber of cuticular scales in a 100micron field of view. Scale frequency isexpressed as a Mean Scale Frequency(MSF). A lower MSF indicates a longercuticular scale. A higher MSF indicatesa series of shorter scales. A literaturesearch revealed that an MSF for woolranges from 10-12, depending onbreed, 6-8 for de-haired cashmere, and6-7 for de-haired mohair.3 Some care isrequired to achieve consistent resultsusing IWTO-97 methodology. A varia-tion of 10-30% was possible. dependingon where you counted the scales. Wooland huacaya alpaca could easily becounted on the edge of the fiber sam-ple, but suri – which has virtually noscale height – was most easily counteddown the center of the fiber. UsingAdobe Photoshop to open the micro-graphs, huacaya scales could be count-ed viewing the micrograph at 50%actual size, while suri required viewingat 66-200%. We standardized oncounting scales on the left edge of thesample viewing it at 100% actual size.

Very precise measurement of fiberdiameter, scale length, height, and scaleedge angle, can be accomplished usingdigital imaging software. We triedScion Corporation’s Image for Windows,and Image-J software, which is availableas a free download by the NationalInstitutes for Health (NIH). The NIH

The golden fleece! Fiber samples have beencoated with 20-30 Å of pure gold prior to beingviewed by the Electron Microscope.

160 ■ Alpacas Magazine

Dr. Chris Davitt, PhD, with sample dishes infront of her Scanning Electron Microscope inPullman, Washington.

A 13mm long fiber sample is attached to theSEM stub with electrically conductive tape.

Image-J was much more stable on ourWindows XP platform than theMacintosh-based Scion Image, which isan industry standard in SEM labs.

Surface SEM of suri alpacas was firstconducted in the United States in 1998at the University of Idaho at Moscow,by Suvia Judd and Deborah Berman.2

Judd and Berman analyzed two fibersamples and hypothesized a high corre-lation between scale length and luster.SEM of South American Camelid (SAC)fiber has been published at least threetimes prior to Judd and Berman in 1988(Phan), 1996 (Antonini et al), and 1997(IWTO). Antonini’s study is particularlyimportant. Extracting the data on justPeruvian suris from that of ArgentineSAC with “Lustre” established thatPeruvian suris have a MSF of 7.5. Thisis equal to cashmere. The 1997 IWTOstudy of specialty fibers is one of manywhich have tried to find an economicalmethod of testing large quantities ofcashmere for purity. The very low MSFof cashmere, compared to wool, makesSEM an effective tool to identifyblended shipments of cashmere.

The author’s sample of suri and specialty fibers is the largest domesticanalysis conducted to date.

Sample Preparation

Dr. Davitt prepared the fiber sample bycutting the center 13mm (½ inch) outof the locks. Dr. Davitt believed usingthe center of each sample was prefer-able to one close to the skin or near thetip, and I concurred. The fiber sampleis mounted on an aluminum disc withelectrically-conductive tape andscrewed into place on the sample dish. A SEM dish holds twelve 13mm sample discs.

The fiber samples were placed in a vacuum chamber and coated with 20-30 angstroms of pure gold. A highvoltage is applied in a vacuum chamberto a bar of 99.9% pure gold, whichcoats the fiber samples with a plasma of gold ions. The SEM is actually visualizing the gold ions that coat thesample, not the fiber sample itself.Platinum and aluminum can also be used to coat the samples.

Commercialization of SEM?

A used SEM costs USD$50,000. Add-ing the other peripherals can raise theprice to USD $250,000. Training toprepare the samples and take micro-graphs would take at least a term ofcollege-level classes in statistics andimaging software. SEM is a powerfultool that can help breeders determinethe inherent luster of their most valuableanimals. But it is unlikely that SEMwill ever be as common a test procedureas laser scanning for determining AFD.The cost of the equipment is just toogreat. Cuticular scales can be visualizedby an inexpensive optical microscopeby coating the fiber samples with a lac-quer that enhances the contrast of thescales. While sample preparation maybe as time-consuming as SEM, this is amuch less expensive method of docu-menting MSF. Unfortunately, the lac-quer greatly exaggerates the cuticularscale height of the sample. Since SACfiber has a very low scale height, SEMis still the preferred method of analyz-ing surface scale structure.

Luster

Luster is the primary, and probablythe only, reason the textile industrypurchases suri alpaca fiber.

Suri can be used in many of the sameapplications as silk and cashmere. It isfrequently blended with merino, silk,or cashmere to add luster to the fabricused in men’s suits.

“Luster is strongly associated with mo-hair, based on its relatively large sur-face cuticle scales and low cuticle scaleedge height relative to merino wool.” 3

The end-use of suri alpaca fiber is sub-stantially different than Baby Huacaya,which can compete for fineness withsome grades of merino wool. Suri ismore likely to be used in a semi-worsted or woolen manufacturingprocess and huacaya in a worsted yarnlike merino. Our study of eight hua-caya alpaca samples demonstrated thathuacaya breeders can selectively breedfor enhanced brightness in their fleece.A reduction in MSF of 25-30% shouldbe relatively easy to accomplish.

Spring 2006 ■ 161

Luster is the primary reason the textile industrypurchases suri fiber. This brushed suri coat byBeatriz Canedo Patiño is made from 100% suri.

A bright fleece like this has a fan-shaped ratherthan a twisted lock.

Lock

It would be premature to assign valueto one lock type over another until amore thorough SEM is made of surialpaca fiber.

The authors had hoped to identifysome correlation between lock types,which naturally occur in the suri alpaca,and the length of scale which is highlycorrelated to luster. Our sample sizewas too small to conclusively identify atrend. However, it is probably accurateto say that a straighter fleece is morelikely to have a very low MSF than atwisted lock. A literature search suggeststhat some lock types may actually inhib-it perceived luster more than others.

For instance, a high frequency ofcrimp in cashmere does adversely affectthe perceived luster of the fiber.

“Perceived luster of wool is affected bystaple structure and fiber curvature.Low curvature in wool allows the fibersto more closely align. As Khan (1996)reported, if the staple crimp form inwool was ‘planar’ (sinusoidal asopposed to helical), such wool wouldhave a high luster. On this basis, theuse of perceived luster as an aid in theclassifying of cashmere may be con-founded by different cashmere fibercurvature. Thus luster of cashmereshould be assessed on manuallystraightened fiber to minimize anyeffect of fiber crimp.” 4

Applying McGregor’s findings to surialpacas may suggest that a tightly twist-ed (helical) lock may diffuse light morethan a straighter fleece. It may alsomake the handle feel coarser than itactually is. Analysis of the Baby Camelsupports the notion that crimp and/ora twisted lock may reduce luster inSAC fiber.

“Wool with larger fiber crimp ampli-tude is associated with softer handle.Wools exhibiting a coiled (helical)crimp configuration receive harsherhandle scores than wools with sinu-soidal (wavy) crimp configurations.” 3

It is probably fair to say that thedefacto breed standard in AOBA sanc-tioned shows puts a much greater

emphasis on an “independent lock,”than any other characteristic, includingluster or conformation. Our SEMstudy did not identify any correlationbetween a twisted lock and luster. Ifanything, they may be inversely pro-portional. In this study, the relativelystraight fleeces, and those with a nar-row straight lock usually had a longerand lower scale height than animalswith twisted locks. Suri breeders, theAOBA Show Committee, and theJudges Training Committee shouldkeep abreast of SEM technology as itapplies to suri alpaca fleeces. It wouldcertainly be premature to favor onelock type over another until the resultsof TSN 100-fleece study is completed.

Scale Height

The scale height of suri alpaca fiberwas almost impossible to measure,even with digital imaging tools. It isessentially a mono-filament, like silk.

The height of scale on suri alpaca fiberwas almost impossible to measure, evenwith digital Image-J software. This char-acteristic is probably as significant to thetextile industry as suri’s very low MSF.The scale height of both suri and hua-caya fiber has not been accurately report-ed in previous scientific literature, whichis probably due to the fact that an opti-cal rather than SEM has been used.

“The smoothness of alpaca and mohaircompared to wool is due to the scalesbeing around half the height of wool ataround the same micron.” 10

While Mr. Sporle made an importantand accurate observation about thehandle of alpaca and mohair, scaleheight is much lower than “one-halfthe height of wool.” In our study, itwas one-tenth the height! Huacayafiber had slightly taller scale heightthan suri, but was still typically under3/10µ. The most highly evolved SoftRolling Skin (SRS) merino fiber likethat analyzed in this study had a scaleheight of 3-4 micron, and most merinois <8 micron. Bruce McGregor, in theAustralian Farm Journal 2003,explained the importance of scaleheight and length when he wrote:

162 ■ Alpacas Magazine

Twenty-one of the 35 suri samples were fromwell locked individuals like this. A twisted lockprobably does reduce luster compared to astraighter fleece.

MSF 4.6. Second and third fleeces had a verylow MSF. Pictured is a 3 month re-growth on"Rewinds" third fleece. There is no reason notto show shorn suri’s in AOBA halter classes!

“The greater the directional frictioneffect due to the wool fiber cuticlescales, the harsher the handle.” 3

This opinion is further enhanced byJ.E. Watts and Janie Hicks:

“Fine cylindrical fibers have low scaleheight. When these fibers are also long,the fiber scales will be long as well asflat. The combination of long flat scalesimparts a smooth or silky feel to thewool.” 5

Because wool has such a high scaleheight compared to suri, whether anindividual fiber is round or elliptical isprobably not as important to alpacabreeders as it is to merino producers.

Round or Ellipical Fibers?

While Watts’ and Hicks’ SRS system isnot universally accepted, their researchstrongly suggests a correlation betweenhandle and a round rather than ellipti-cal fiber. When I showed these SEMmicrographs to Ian Watt, he soonfound fibers that he believed were ellip-tical rather than round. These are easilyidentified by two thin grooves runningparallel to the direction of the fiberwith a rounded segment between them.

About a third of suri and huacayasamples had one or two grooves run-ning the entire length of the fiber.Most were completely round. Nollama, guanaco or camel samples

appeared to be elliptical. One vicunahad distinct grooves. Since there werefrequently two fibers in each micro-graph, and occasionally three, we ana-lyzed nearly 200 suri and huacayafibers for roundness. The elliptical sam-ples tended to come from suris thatwere well locked, but not exclusivelyso. More research will be needed todetermine what, if anything, this differ-ence in fiber profile means. TEMand/or skin biopsies would be a moreaccurate way to determine if a hair fol-licle is round or elliptical.

To an end-user, the very low scaleheight of suri is probably of muchmore significance than if an individualfiber is round or elliptical. You canappreciate this for yourself by feelingthe “hand” of a neck scarf made frompashmina (cashmere) or merino woolof similar AFD.

Transmission Electron

Microscopy (TEM)

TEM can also be used to visualize thecortex of a fiber sample. The percent-age of orthocortical and paracorticalcells in the cortex is important. Wool isbilateral, with both orthocortical andparacortical cells, which is responsiblefor its crimp.9 Cashmere with moremesocortical cells, has a highermicrofibril packing density than woolof the same diameter, so it has less curvature. Optical analysis of suri fiber

Spring 2006 ■ 163

The low scale height of suri can clearly be seenin this micrograph which shows three damagedcuticle scales. Notice how thin they are!

Elliptical Suri Fiber – MSF 5.25. About 30% ofthe alpaca samples had one or two grooves running along the surface of the fiber, indicatingthey are probably elliptical rather than round.

Compare the lack of scale height of this surisample to that of merino or cashmere (see photos on p. 168-169).

The scale edge of suri is so indistinct that itwas difficult to measure even with digital imaging tools.

suggests it is not bilateral, and thiswould help explain why it is such astraight fiber. It is much more timeconsuming to prepare samples for TEMthan SEM, and therefore more expen-sive. TEM will probably require indus-try support from ARF, AOBA or TSN.

Statistical Analysis

“7.0 Scale/100 micron seems to be adistinctive parameter for suri.” 6

Dr. Davitt supplied the author with1.7 megabyte, 8 bit gray scale, TIFFfiles captured with a Scion Grabbercard from the SEM. Two micrographswere taken of each sample. One tothree fibers could be measured in eachmicrograph. These black and whitemicrographs had remarkable brightnessand contrast. I used the NIH Image-Jsoftware to measure the diameter offiber samples, and using the angle tool,the angle of scale perpendicular to thefiber. There was an obvious differencebetween huacaya and suri samples. Surisamples were typically less than 45degrees, while huacaya and llama werecloser to 70 degrees. Both scale edgeangle and MSF can be used to identifysuri from huacaya or llama fibers.

Using modified IWTO-97 method-ology, the MSF of our suri samples was 6.15 scales per 100µ. This is about 20% superior to cashmere or the Peruvian suris in Antonini’s study.Scales were frequently as long as thefiber was wide. There did not appear to be a relationship between scalelength and fiber diameter as reportedin the literature for other specialtyfibers. This is good news for suri breed-ers, since selecting for fineness shouldnot adversely affect luster. The averagescale length was 16.25µ. Antonini concluded that;

“7.0 Scale/100 micron seems to be adistinctive parameter for suri.” 6

While highly-evolved suris like thoseI analyzed have an even lower MSF, ourdata certainly supports that assertion.Separating out the Phase 2 data whichwas taken from a uniform collection site,the difference between our suris andAntonini’s was even more significant.

164 ■ Alpacas Magazine

MSF 10.0. The highest frequency suri had alower MSF than the mean of all huacaya samples.

MSF 5.O Suri is nearly as bright as mohair buthas a much lower AFD.

MSF 6.5. The MSF of suri alpaca fiber in thestudy of 35 suri alpacas was 6.15/100 micron.

MSF 7.0. The suri sample in this image has alower MSF than cashmere.

MSF 5.75. Suri breeders need to reduce the AFDof their clip to be competitive with other specialtyfibers. This sample is both fine (<19µ) and bright.

The finest suri analyzed may be showing signsof a “wool break” on the far right. AFD was just12-16µ.

The number of scales per 100µ varied depend-ing on where you counted them. Suri guard hairhad a MSF of 8.65 with a range of 4-13.

■ MSF 5.5 2nd & 3rd fleece■ MSF 5.8 virgin fleece■ Range 3-10.25■ MSF guard hair 8.65■ Range 4-12 ■ Phase 1 & 2 combined MSF 6.15■ Substantially lower than Antonini’s

Peruvian data (7.5).

MSF should not be disassociatedwith scale height. When the scaleheight is impossible to measure, andyou have to zoom in on the micro-graph to 200% actual size to even visu-alize a scale edge, you have what isessentially a mono-filament like silk.Several of the suris with a relativelyhigh MSF of 7-8 were so smooth thatit still had a very slick, cool handle.

Guard Hair

Primary hair follicles (guard hair) ofthe Camel Family had a distinctly dif-ferent scale pattern than the secondaryfibers. With the exception of vicuñas,they were usually much stronger (high-er AFD) and had an extremely highfrequency of scale. These primary fibershad extremely low scale height, anddespite their AFD and scale frequencyprobably had good luster. They wereuniformly round, rather than elliptical.The scale pattern had what appeared tobe a fractured or very busy “mosaic”appearance. This, combined with theirlarge AFD, made them relatively easy

to identify. There was a significant dif-ference between the MSF of guard hairon suris between Phase 1 & 2. Thiswas probably due to the uniform col-lection site used in Phase 2 and the factwe intentionally included some samplesthat were “hairy” or had a strongerAFD in Phase 1. It is probably accurateto say that SAC typically have guardhair with about 16 scales/100µ.

Huacaya

Eight huacayas were evaluated, fivemales and three females. Samples werecollected from virgin fleece. These werehighly-evolved huacayas, with 2.0 to3.0 crimp per cm, clearly definedbundling, and two were unusuallybright. AFD was 16-19 micron. Whileour suri samples were randomly select-ed, based on fleeces or Yocum-McCollsamples we had on hand, these hua-cayas were hand-picked to represent adense, bundled, extremely fine, or verybright fleece. Huacaya samples had anMSF of 11.0, which is similar to wool,with a range of 8.0-12.0 MSF. Thiscorrelates to an average scale length of9.0µ. Scale height was greater than suri,but typically <0.3µ, which is ten timesless than the most highly-evolved meri-no wool. The scale edge angle of hua-caya fibers was about 70 degrees, com-pared to 45 degrees for suri. This maybe responsible for its slightly greaterscale height. Among the huacaya sam-

Spring 2006 ■ 165

Suri alpacas do have guard hair. Here we see a24µ secondary fiber in front of a 40µ guard hair.

Bright Huacaya – MSF 11. Huacaya fiber had ahigher MSF than suri and a 70 degree ratherthan 45 degree scale edge angle.

Fine Huacaya – MSF 11.0. Huacaya sampleswere very fine, seldom exceeding 19µ.

Sheepy Huacaya – MSF 10-12. This is either a16µ guard hair or a very "sheepy" secondaryfrom a fine huacaya. Note how similar it is tothe undercoat of the double-coat llama (seeimages on the next page).

ples, the longest scale length also hadthe greatest amplitude of crimp. Thiswas a bright, well-bundled, “elite”fleece. Primary and secondary fiberswere nearly the same AFD, though theprimary fibers could always be identi-fied by their high frequency, and“sheepy” scale pattern.

Llama

There are three distinct breed-types ofllama raised in the United States. Wetested examples of each, including: atraditional short-wooled, double-coatedccarra llama (32µ AFD and >50%guard hair); eight single-coat llamas(<9.5% coarse fibers and 21µ AFD);and 11 SLA-registered suri-llamas,most of which had pedigrees that wereseveral generations long. (They are nothybrid alpaca x llamas).

The surface structure of a ccarra fibersample looked almost identical to hua-caya samples with an MSF of 11.5.However, their primary hair follicles(guard hair) had a much greater AFD,and was routinely >50 micron.Secondary fibers averaged 24 micronsand looked very “sheepy.” The llamas’guard hair had an MSF of 16.5 with arange of 14-19. The surface structureof a double-coat llama shows a muchgreater similarity to huacaya fibers thansuri. This may be of interest toarchaeozoologists, and could have sometaxonomic significance.

Secondary fibers of single coat llamawere very suri-like with an MSF of 8.0and a range of 7.5-10.0. These highly-evolved 21µ llama fleeces are indicativeof what is shown in Alpaca and LlamaShow Association (ALSA) medium-and long- wool halter classes. Startingfrom very humble beginnings, it hastaken llama breeders 30 years (10 gen-erations) to develop from a double-coat

to a single-coat fleece of this quality.These llamas had a scale height lessthan or equal to huacaya alpacas.Handle could be either warm or cool.

The suri-llama samples had an MSFof 7.0, which is slightly superior toAntonini’s Peruvian suri alpacas. Six ofthe 11 donors were top ten WorldFuturity suri-llama champions. Alpacabreeders should take note of howquickly llama breeders have been ableto improve fiber characteristics.Huacaya breeders should be able to addbrightness to their fleeces relativelyquickly using SEM, and suri breedersshould be able to simultaneouslyreduce AFD while increasing luster.

Baby Camel

The two Baby Camel samples (Bactriancamel) were imported from China.They had been de-haired to less than2% coarse fibers (>30µ). The BabyCamel was very fine, <18 micron, andhad an MSF of just 6.875, with a rangeof 6.15-8.0. It’s resistance to compres-sion felt less than huacaya, but was stillvery “lofty” compared to suri. It had awarm, rather than cool, handle. Thefrequency and amplitude of crimp wasnot visible since all of its fibers hadbeen aligned into a roving. It did notappear to be a particularly bright fleece,yet it had an MSF similar to suri,which had not been washed or de-haired. Plucking a string of fiber out ofthe camel’s roving and wrapping itaround our fingers revealed that it wasmuch brighter than our first impres-sion. (This straightens the fibers).

As McGregor found with cashmere,crimp does appear to reduce luster. Thewarm handle of Baby Camel may alsosuggest that scale height and crimp areas important as MSF. The scale heightof camel did not have an abrupt edgelike wool or cashmere, but was measur-able with digital imaging tools and wassimilar to huacaya. Dr. Davitt notedthat the specialty fibers like cashmere,silk, SRS merino, and Baby Camel,were definitely more “fly-away” thansuri. She even wondered if they would“stay on the SEM stubs when the elec-tron beam hits them!”

166 ■ Alpacas Magazine

Guard hair from a short-wool, double-coat llamahad a MSF of 16.5 with a range of 14-19. Thissample is 50 micron AFD, which is not unusual.

Single-Coat Llama – MSF 8.0. Single-coat llamashad a MSF less than SRS merino wool!

The undercoat of the double-coat llama is muchfiner (24µ) than its guard hair. The undercoat issimilar to a "sheepy" huacaya alpaca.

■ De-haired baby camel■ Low MSF (6.875)■ Higher scale height than suri■ Crimp reduces luster & handle■ Brighter if straightened

Suri breeders should be proud of thefact that a random sampling of Boliviansuri alpacas equals or exceeds the MSFof washed and de-haired Baby Camel.

Guanaco

Dr. William Franklin, one of theworld’s leading authorities on guanacosand vicuñas, supplied wild-caught gua-naco samples from the Falkland(Malvinas) Islands. Four of the samplescame from the back near the top line,and one from two males who were bit-ing each others’ necks. Four were red-dish-brown to fawn in color and onesample was white, probably comingfrom high on the side. This was truly arandom sampling! ■ 5 samples from the Falkland Islands■ MSF 11.0■ 10-60 micron AFD■ Lower MSF than vicuña■ Distinct guard hair with undercoat

Based on the similarity of their inci-sors, the guanaco is probably closelyrelated to the llama, whereas the vicuñais closely related to the alpaca. TheMSF of these authentically-pure guana-cos certainly supports a close relation-ship between the guanaco and the dou-ble-coat ccarra-type llama. The under-coat of the guanaco samples was veryfine, and some fibers were as small as10 microns! However, the guard hairwas much more prevalent than any ofthe other breeds or species of SAC, andseveral 60-micron guard hairs woulddominate nearly every micrograph,making it difficult to isolate the much-

finer secondary fibers. It would beinteresting to analyze de-haired guanaco roving.

Vicuña

We analyzed three wild and threedomestic vicuñas. Dr. Franklin sup-plied the vicuña samples from thePampa Galeras National VicuñaReserve in Peru, and Dr. Toni Cottonthe domestic vicuñas from Jack andMiriam Donaldson’s farm in Findlay,Ohio. The wild vicuñas were twoadults and one cria. Samples were col-lected from center/middle back, topcenter neck, and center of back withguard hair intact. The domestic sam-ples were from adult males, and hadbeen de-haired by hand. Staple lengthwas 40-50mm. The providence of cap-tive vicuñas is not as likely to be aspure as wild vicuñas, especially inEuropean zoos where South Americanexhibits will frequently house all themembers of the SAC together.However, these samples had largelysimilar MSF and AFD. ■ 6 samples (3 wild-caught)■ Range 8.5-15 scales/100 microns■ MSF 12.1875■ 12-15µ AFD■ Very uniform

The vicuña samples were among themost interesting of the study. Theyhad a higher MSF than either hua-cayas or llamas, and the differencebetween primary and secondary fibers

Spring 2006 ■ 167

Baby Camel – MSF 6.875. Crimp greatly reducedthe luster of this de-haired roving. Straighteningthe fibers showed its inherent luster.

Vicuña – MSF 12.875. The vicuña samples had ahigher MSF than llama, guanaco or huacaya.AFD 12-15µ.

Guanaco fiber was characterized by 60 micronguard hair and very fine undercoat. The guardhair is in front of the 10-12µ undercoat .

was difficult or even impossible todetermine by SEM. This may be onereason the MSF was higher, since withall other members of the SAC, it waspossible to count primary and second-ary fibers separately. This will beamusing to those of you who haveever de-haired a vicuña fleece. (Thoselittle red guard hairs are easy to see buthard to pluck out!) The MSF andAFD of the vicuña samples were themost uniform of all SAC tested. Thisis a significant observation, consider-ing the random nature of the wild-caught samples. A follow-on analysisof vicuña fiber with SEM shouldprobably analyze guard hair separatelyfrom the under-coat.

HYBRIDS:

First-Generation Suri x

Huacaya Cross

Kenneth Madl of Aviana Farms, sup-plied three colored suris from his herdwhich is agisted in Strathbogie,Victoria, Australia. Two were black andone silver-grey, 11 months old. Twowere F1 suri’s with 2-3 generations ofsuri x huacaya crosses in their pedigree,and one was a Back Cross 1 with twogenerations s x h and one s x s. Two ofthese went on to win their classes at the2005 Australian national alpaca show.Over 3,200 suri alpacas were importedinto the United States from Peru,Bolivia, and Chile between 1991-98.With such a large population to startwith, the vast majority of Americanbreeders do not cross huacaya and surialpacas. Many fewer suris were import-ed into Australia, and as a result it is acommonly-accepted practice in thatcountry to cover a huacaya female witha suri male. ■ Three samples from Australia■ Two true black, one silver-gray■ Range 6.25-11 scales/100µ■ 8.4 MSF (intermediate between

huacaya & suri)

As the photos indicate, these alpacasare phenotypically suri and would notbe out of place in most colored surialpaca herds in the United States.While it is a small sample size, the

MSF is intermediate between a huacayaand suri, which may or may not be dueto their huacaya ancestry. (The suriwith lowest MSF had the greatest per-centage of huacaya pedigree). The scaleedge and height on 2 of the sampleswas more distinct than most suris, par-ticularly one of the blacks.

Paco-Vicuña

A paco-vicuña is the offspring of analpaca and a vicuña. A single paco-vicuña sample was evaluated from theherd of Phil and Chris Switzer in EstesPark, Colorado. Five fibers were ana-lyzed in two micrographs. The provi-dence of this animal could not be veri-fied, but this breeding male had manyvicuña-like characteristics, includingbody size, weight, bi-coloration, staplelength, and personality. It was an in-utero offspring out of an importedChilean huacaya. ■ One sample■ Range 11-17 scales/100 microns■ 13.6 MSF■ 13-24 micron AFD■ Higher MSF than huacaya or vicuña

This animal had the highest MSF ofthe series, which probably goes a longway toward proving its vicuña pedigree.The uniformity of AFD was not nearlyas pronounced as a pure vicuña.

OTHER SPECIALTY FIBERS:

Soft Rolling Skin (SRS) Merino

As anyone who has ever shopped for anexpensive man’s suit can tell you, ultra-fine merino wool should probably beclassified as a “specialty” fiber. Even anuntrained hand can distinguish the dif-ference between a 16µ wool and morecommon fabric. When merino isblended with cashmere or suri, thehandle improves even more. The twoSRS merino samples we tested had aMSF of 8.6, which is exceptional forwool. These two- and three-year-oldmerino rams had a scale height of just3-4 microns, half that of most sheepbreeds. However, the abrupt scale edgewas easily measured. These wereremarkably well-bundled fleeces, witheach lock measuring less than ½ cm

168 ■ Alpacas Magazine

Paco-Vicuña – MSF 13.6. Paco-vicuñas had thehighest MSF in the study, and a much greaterrange of AFD than the pure-bred vicuñas.

SRS Merino – MSF 8.6. While half the scaleheight of other sheep breeds, SRS merino is still more than ten times greater than huacayaor suri alpaca.

Suri x Huacaya Cross – Average MSF 8.4. ThreeF1 or BC1 suri x huacaya crosses from Australiawere analyzed.

with 6-7 crimps per cm. (This is twoto three times the curvature of a typicalhuacaya). The two samples were pro-vided by Ian Watt of Moro Bay,California.

Cashmere

SEM has demonstrated that suri fiberis most similar to cashmere andmohair. These goat fleeces must be de-haired, but have an MSF and surfacestructure that is much more similar tosuri than huacaya or wool. While hua-caya breeders can learn much fromresearch based on merino wool, suribreeders should probably look to otherresources.

Two cashmere samples were ana-lyzed. One sample was collected in thegrease from a domestically raised doe.She had an MSF of 7.6, scale height of1-2µ, and 17µ AFD. We also tested awashed and de-haired cashmere samplewhich was imported from China andsupplied to us by Angus McColl. Mr.McColl said that white Chinese cash-mere was finer than other colors. It wasvery fine, at 14.2µ AFD. Since it hadbeen washed, it was also very bright,but had an MSF of only 8.625, aboutlike SRS merino. Its range was 7-10scales per 100 microns, and both sam-ples were within the range of AFD andMSF reported in a literature search forcashmere.3, 7, & 8

Over five-thousand metric tons ofcashmere are sold each year, comparedto just 550 tons of suri. By verifyingthe low MSF and scale height of suri

with SEM, the alpaca industry shouldbe able to substitute suri fiber for manyapplications that currently use cash-mere or mohair.

Mohair

We evaluated a single sample ofimported fawn mohair. Most mohairproduction comes from South Africa.Mohair is brighter than cashmere, andsubjectively, appears to be as bright assuri. This sample had been washed.Even though mohair is very coarse(typically <32µ), it still had a lowerMSF than suri. This is largely offset byits much greater AFD and more dis-tinct scale edge. Goat fleeces must bede-haired, but are microscopically moresimilar to suri than wool. Seven-thou-sand tons of mohair are sold each year,indicating the great potential that surifiber does have. Suri alpaca can certain-ly be marketed as having similar lusterto mohair, while being much finer. ■ One sample■ South African mohair■ Mohair is from a goat like cashmere■ High luster■ MSF 4.25■ Distinct scale compared to suri■ 32 microns (coarse)

Angora Rabbit

Angora is a single-coat fiber that doesnot need to be de-haired. Suri breederKathleen Cullen gave me a sample ofwhite Angora from her flock ofGerman rabbits. White Angoras arelarge-bodied and suitable for harvestingtheir fiber. The cuticle scale pattern ofAngora was unlike any other fiber sam-ple. It can best be described as an“other-worldly” spiral pattern.

Spring 2006 ■ 169

Mohair – MSF 4.25, AFD 32µ. Mohair is brightbut coarse.

Angora rabbit – MSF 10.6. Angora is a singlecoat fleece that does not need to be de-haired.It is has an unusual "spiral" scale pattern.

White Chinese cashmere, MSF 8.625, 14.2µ AFD.

SRS Merino Fleece

Mohair Fleece

170 ■ Alpacas Magazine

MSF was not related to AFD as it appears to bewith other specialty fibers. This is good newssince it suggests you can select for enhancedluster and reduced AFD simultaneously.

■ Single coat fleece■ 11-15 micron AFD■ Spiral scale pattern■ Range 10-12 scale/100 microns■ 10.6 MSF

Angora is so fine and without guardhair that we could easily measure threefibers per micrograph. The deepgrooves between the spiral-shapedscales probably reduce the handle ofAngora in the same way that an abruptscale edge effects wool. The scale edgeangle is similar or greater than huacaya.

Bombay Silk

Imported Bombay silk had an AFD of10-12 micron. “Bombay” silk comesfrom the mulberry leaf eating silkworm. One-hundred-ten-thousandtons of silk are sold each year, makingit the dominant luxury fabric in theworld. Since it is not a mammal, silkdoes not have a cuticle scale structure.Silk is reeled together from 4-20 indi-vidual filament ends to make a singlemonofilament that adheres togetherdue to the gummy texture on the sur-face of the filament. Silk is washed inwarm water to separate the filament

from the silk worm’s cocoon, and againafter it is wound. It was very bright.Even at 1,000 X, it was not possible tosee where the filaments had been reeledtogether. It appears that silk is the onlynatural fiber which has a smoother surface structure than suri.

Challenges and

Recommendations

SEM of suri fiber is a fortuitous con-vergence of research and marketing.

Suri alpacas have a lower MSF andmuch lower scale height than cashmere,but no one knows it! To secure theirplace as the natural fiber with thegreatest combination of fineness andluster, AOBA judging standards needto prize luster as highly as an independ-ent lock structure. And we believe thisneeds to occur in the show ring, notjust on the score card for a fleece com-petition. Additionally, we recommendthat the Alpaca Research Foundation(ARF) help sponsor research on surifiber. (The Suri Network already is).Clearly, AOBA and the Alpaca FiberCooperative of North America(AFCNA) both need to make suris’unique characteristics an integral part

IWTO methodology views fiber samples at 600 Xmagnification which allows 5-6 fibers to beanalyzed at a time. Dr. Davitt increased magnifi-cation to 1,000 X so suri scales could be visual-ized more easily.

The authors would like to thank the following friends and colleagues for their assistance: ■ Dr. William Franklin, who supplied us with wild-caught guanaco and vicuña samples

from Pampa Galeras and the Falkland Islands. ■ Dr. Toni Cotton of Findlay, Ohio, who provided the domestic vicuña samples.■ April and Richard Angotti of Benvenuti Farms in Bend, Oregon, who

provided the highly-evolved huacaya samples.■ Sherry Sheridan of Aloha Farms in Bend, Oregon, who provided the

double-coat llama sample.■ Susan Tellez for her foreign language literature search of SEM of the SAC. ■ Kenneth Madl, who provided us with many publications on contemporary wool and

cashmere research from Australia, and his first-generation suris from suri x huacayacrosses.

■ Ian Watt, for help in identifying round versus elliptical fibers, and for providing thehighly-evolved SRS merino samples.

■ Our local wool guild, the Textilians, who provided us with cashmere, silk, and BabyCamel (Bactrian camel) samples.

■ Angus McColl, who provided the white Chinese cashmere. ■ Dr. Kathleen Cullen of Spokane, Washington, who provided the white Angora.■ Chris and Phil Switzer of Estes Park, Colorado who supplied the paco-vicuña sample.■ Dr. Chris Davitt, WADDL, who prepared and scanned the fiber samples and tutored

me on the use of imaging software.

Acknowledgements

Bombay Silk – AFD 10-12µ. Silk does not have a cuticle scale since it is reeled together from4-20 filament ends from the silk worm.

of their marketing and advertising cam-paigns. The low scale height of huacayafiber compared to wool can also beemphasized! To be competitive in theluxury fiber market, suri breeders needto reduce the AFD of their clip.Fortunately, there was no correlationbetween luster and AFD in either thisor Antonini’s study, and it should bepossible to select for both characteris-tics simultaneously.

Cheryl and I have been raising alter-native livestock for over thirty years.As a result we usually take a macroview of the issues which face ourindustry. We have learned that if youdo not breed for something, you loseit. Whether this is maternal ability(milk production), thriftiness, confor-mation, or luster, if you take a charac-teristic for granted, you will lose it.The results of this research are bothexciting and sobering. For instance,there is certainly no reason not toshow shorn suris in AOBA halterclasses, due to the close correlationbetween a virgin, 2nd, and 3rd fleecesMSF. On the other hand, we are probably not using some of our bestmales, those with the greatest lusterand lowest MSF.

Summary

It would be premature to assign valueto one lock type over another until amore thorough SEM study is made ofsuri alpaca fiber. Like many scientificenquiries, this one has raised as manyquestions as it has answered. Theauthors have scanned enough fibersamples to determine that suri alpacafiber is measurably different in scalelength, frequency, edge height, andedge angle from wool and huacayafiber. It is probably most similar tocashmere and mohair but with an evenlower scale height. The “take home”lesson from this SEM study is thatwhat is truly unique about suri alpacafiber is its luster. This should be keptforemost in mind in any future discus-sion of possible breed standards, end-use, or show standards. ■ Suri and huacaya fleece can be differ-

entiated by SEM

■ The “long smooth scale”™ of surifiber averages 16.3µ long

■ Huacaya samples averaged 9.0µ■ This equates to a MSF of 6.125/

100µ suri and 11.0 MSF/100µhuacaya

■ MSF was not correlated to fiberdiameter.

■ A low MSF was not correlated to atwisted lock and may be inverselyrelated

■ Scale height on both huacaya andsuri is much less than previouslyreported.

Andy and Dr. Cheryl Tillman funded thisresearch on suri alpaca fiber. The Tillmansimported the influential colored suri alpacasfrom Hacienda Acero Marka near La Paz,Bolivia in 1996. Andy has raised llamas since1975 and alpacas since 1982. He is a memberof The Suri Network research committee.Cheryl Tillman has been a camelid veterinariansince 1985. She sits on the board of directorsfor the US Animal Health Association, serveson the AOBA/ARI Government & IndustryRelationship committee, and has reviewedresearch proposals for Morris AnimalFoundation for many years. They can be contacted at andy@tillmansranch.com.

REFERENCES1. College Physics, 4th edition, Franklin Miller Jr., pg. 6592. Purely Suri magazine, volume 2, 2004, Suvia Judd and Deborah Berman, pg 183. “Is Merino Wool a Luxury Precious Fiber?,” Bruce McGregor, Australian Farm Journal,

July, 20034. “Influence of Nutrition, Fiber Diameter, and Fiber Length on the Fiber Curvature of

Cashmere,” B.A. McGregor, The Australian Journal of Agricultural Research, 2003 5. J.E. Watts and Janie Hicks: The Soft Rolling Skin® (SRS)® Breeding System for Alpacas.6. “SUPREME-Project: Cuticular Cell Mean Scale Frequency in Different Types of Domestic

South American Camelids (SAC)”; M. Antonini, E. Frank, M. Gonzales, F. Pierdominici, S. Catalano, M.V.H. Hick, & F. Castrignano; European Symposium on SAC, University di Camerino Press, Matelica (MC) Italy, pp. 142-152 (1996)

7. IWTO DTM-XX-97: “Quantitative Analysis of Blend of Wool with Specialty Fibers by ScanningElectron Microscopy,” International Wool Textile Organization (1997)

8. “Characterization of Specialty Fibers by Scanning Electron Microscopy”; Phan K.H., Wortmann F.J., Wortmann G., and Arns W.; Shriftenz. Dtsch. ; Wollforxchungsinst; p. 106 (1988)

9. Properties and Performance of Goat Fibre, JD Leeder, BA McGregor and RG Steadman (1998)10. Elite Natural Fibre, Grower Expectations & Processor Requirements, Peter J. Sporle, 199411. Alpaca: Large Scale Industrial Perspective on Fiber Processing, Juan Pepper, 2005

Spring 2006 ■ 171

A fan-shaped lock and narrow flat lock may beunder-appreciated for luster.

It would be premature to assign value to onelock type over another until a more thoroughstudy of SEM is made of suri alpaca fiber.