go to index page
go to genetics 1
go to genetics 3
 

Applied Genetics , a Genetic Discussion, as it relates to the Burmilla
The reader should note that this page is a compilation and consolidation of the previous genetics material . updated feb 09 cinnamon and fawn

These discussions assume that the reader has some knowledge of, and is interested in, broad based cat genetics. We do not spend any time on the mathematics of homozygotes, heterozygotes and predictions of checkerboards. We will not try to teach the subject.  We rather point out the characteristics of the breed, as they appear and as we see them. As such, proofs can only be conclusive when dealing with recessives.
We are not always correct and do not expect to be. Sometimes we must amend our conclusions in light of subsequent breeding, and increased understanding, and we hope that in putting these opinions forward that they will encourage other breeders to come forward with their observations.
There are many good web sites on the Internet on "cat genetics". see http://www.kolumbus.fi/sarakontu/link/catlinks.html

 For our text references we use Gloria Stephen's book Legacy of the Cat edit #1and 2, and Robinson's 'Genetics for Cat Breeders and Veterinarians' edit #4. We recommend the beginner to Gloria Stephens's book Legacy of the Cat. It has more information packed into a few pages, with explanatory pictures, than any other text we have found. We use both books bouncing from one to the other.
Gloria also has a good web site with many links. (Key word her name on a search engine.) A very good article by Breton /Creek is posted on the Net Pets site. That same article is posted on various sites. It has, what I suggest is a, formatting problem on its second chart that created some confusion for me, but the same article is posted on Ms. Stephens's site in its original form. This cleared up my misunderstanding.
Take warning, this subject is not for the faint hearted or those "clone breeders" who spent their time discussing head type and genetically manage little else. We struggle with the subject as those readers of our web site can attest to.
We are dealing with the genetics of cat fur. We must stress this, the terminology is genetic only and the colour names we use are not necessarily related to the terms used in 'the cat fancy'. We claim no expertise in this area and shy away from this minefield. It is often only this aspect of fur colour/pattern that differentiates a "Breed' so that within the Burmilla (Burmese/Chinchilla) gene pool are genetics that are exclusive to many other 'breeds'. The only other breeding programs, we can think of that has this diversity are the Oriental short and longhair and the Abyssinian/Somali.

Genetics of Fur in the Burmilla Program (note capital letters denotes dominance)
In play are A.B.C.D.L.O.Ta.I. Wb The terms used are mainly from Robinsen' s and Stephens

Wide band is new, hypothesized by Robinson’s, and recently accepted by Ms Stephens in her latest edition. We accept it as fact as you see. We are beginning to think of very wide band cats as in the Chinchilla as super wide band (swb), normal/not being SWB. This is our descriptor, as we know of no other. We could use the term Tipped but there does appear to be some assocaition with wide band.
We have discounted Dm, Dilute/non modifier (from Robinson’s 4th ed  p141) after running it for some time. Thus the terms Caramel, Taupe and Apricot are not used. We do not dispute their existance rather we understand these are the minor colour variations one does see (in the solid cats -most particularly the "dilutes") due to the underlying golden /silver variable that becomes clearly visible and is identified in the tabby [agouti] cats.

A Sable Burmese is considered by definition and acceptance to be a non agouti, black, sepia, dense, shorthaired, non orange cat with recognized variation to give chocolate, dilute and orange cats.
[a/a.B/-.cb/cb.D/-.L/-.o/o with possibilities of b/b…d/d..O/O .Ta/- and i?? wb?? swb ??]\
It is generally considered to be most often in its best form to be abbytaby but the breed profile does not address this specifically. Other genes are not commented upon by the breed profile.

Renaissance a Sable Burmese

A Chinchilla Persian is considered to be agouti, black, full colour, dense, longhair, not orange, a Mackerel or Classic tabby thus not abytaby, silver and wideband (Stephens edit #2 pg. 21.)  [A/-.B/-.C/-.D/-.l/l.o/o.ta/ta.I/-Wb/- ]

Christian

Thus we are mixing
a/a.B/-.cb/cb.D/-.L/-.o/o with possibilities of b/b…d/d..O/O .Ta/- and i?? wb?? swb ?? in the Burmese
with
A/-.B/-.C/-.D/-.l/l.o/o.ta/ta.I/-Wb/- in the Chinchilla

You can see the nomenclature we intend to use and should note; the O is sex linked and males will be O/x , and dominants will be as in D/- unless the carrier is known giving D/d.

All first generation Burmillas tend to be similar in appearance with variability around O and the recessives the Chinchilla might carry as in I/i i.e. it had a golden parent.
Thus they are most often:-
A/a.B/-.C/cb.D/-.L/l.o/o (unless noted)Ta/ta.I/i.Wb/wb this being their "genotype"

Andromeda from Christian and Renaissance
 

If a homozygous (no recessives) Chinchilla was crossed with a Platinum/Lilac Burmese (b/b...d/d) the result would be
A/a.B/b.C/cb.D/d.L/l.o/o (unless noted)Ta/ta.I/i.Wb/wb with no physical difference to the cat above (Phenotypically the same) and being
agouti, black, full colour, dense, short, not orange, abytaby, silver, wide band.

If the same Chinchilla was crossed with a Cream Burmese female (d/d and O/O) the result would be
agouti, black, full colour, dense, short, orange in males and tortie in females, abytaby, silver, wide band as below

Venus and Orian from a litter of Angel a cream female Burmese and Christian

It is only when the breeder moves into the subsequent generations that the full coat and pattern variability begins to come into view, and the cat truly develops its Burmese characteristics and conformity.

A Agouti  A/- and non agouti a/a
All cats are Agouti or not

agouti  or not A/-    or   aa non agouti or self - -   the boy on the right a Sable Burmese.

Agouti banding is easily recognized. We will not repeat what the texts have to say on this subject. It is an early evolutionary development as it is seen in a number of animals and thus other characteristics that have evolved more recently key off this gene. I.e. the abbytaby must first be A/-(agouti).  An a/a cat masks (is epistatic) this characteristic thus a Burmese can be abytaby, mackerel, classic and spotted, or not without the breeder ever being aware, as we will see.
The brick nose is related to all this, but I don't know the mechanics. In our gene pool all a/a cats do not have one. It now seems all A/- cats do. Of those, the brick nose can appear as late as 4 weeks after birth, in some cases some agouti cat noses have some ground colour on them. We note also that there is often an area of increases rufousion (redness) in the fur around the brick nose. We can not find a text reference. This is am important differentiation in the "smoke discussion that follows.
(Sidebar O cats don't count, 'cause it’s masked??)
O cats have other differences and I will give the reader a direct quote from Stephens edit #2 pg. 15. "All red or cream-colored cats will show a tabby pattern to some extent...Phenotypically (in their outward appearance) these cats may look like tabbies, but may be genetic solids. ..as a rule of thumb if the red cats chin is "whitish" the chances are the cat is a genetic tabby,..."
From a judging and breeding point of view it is extremely difficult (impossible) to differentiate an A/ red from an aa red (and particularly cb cats that are dd) and can only be determined by heredity or by subsequent breedings.

B Black
All cats are black (B/-)!! (Top cat) Unless they are b/b (as in second from bottom) in which case they are brown.


Colour can sometime be very difficult to identify because as the reader will see there are many factors that reduce, modify, and change colour.
Example

Ecstacy was  a B/- cat . She is a black cat with the Burmese restriction (cb/cb) which makes her a sable. It is the A/- , the golden (i/i) and the cb/cb  along with the long hair (l/l) that in this case deliver the colour that the eye sees. It is the paw pad colour that defines the issue.
 
C Full Colour 
All cats are  C/- full colour/full expression or not. Those that are not are cb/cb Burmese sepia.
The sepia gene is discussed under the title "Pointing". cs/cs is called pointed and is what the Siamese has and is not in our gene pool. The c genes renders the colour manufacture, temperature sensitive so that colour is less intense next to the skin, when the kitten is born, and more intense at the extremities

The sepia effect should be apparent. We know that this gene has varying capability as seen in the Tonkenese and its cs/cb definition. The naming of this intermediate colour effect is as ‘mink’. It is here worth pointing out that Burmese breeders who strive for the very dark cat could be breeding a fundamental error from a genetics point of view

 D dense pigmentation and d/d (dilute sometimes-called maltesing)
The gene results in the clustering of the pigment granules on the hair shaft and allowing a different light reflection so that black , diluted is indigo, sable becomes blue, ,brown becomed caramlee/fawn,  chocolate becomes platinum/lilac, and red becomes cream.(see Stephens)

	black B/-.C/-.D/- is black, full colour, dense at the top left becomes indigo  B/-.C/-.d/d  is black, full colour, dilute
	sable B/-.cb/cb.D/- is black, sepia, dense becomes blue B/-.cb/cb.d/d is black, sepia, and dilute
	brown b/b.C/-.D/- is brown, full colour, dense becomes frost b/b.C/-.d/d is brown, full colour, dilute The reader should here note the term caramel is much used and abused. We use it only as others have and prefer the term Fawn.	not got yet
Lancaster  as a kitten and adult	champagne b/b.cb/cb.D/- is brown, sepia, dense becomes platinum(lilac) b/b.cb/cb.d/d is chocolate sepia dilute   Note As with the reds below the dilute colours are difficult to differentiate from each other in many cases. It is often not clear if a cat is fawn or lilac, blue or  indigo especially if one is looking at a silver based fawn or a golden based lilac, for example. It is this group that most often confuses the naming processes and judges as is seen with the confusion around the term Caramel. It is also this group that has given rise to what we would suggest are false genes i.e. Dilution modifier and B1 or light brown. This is best differentiated by photographing or viewing together.
	AND
	(russet) red boy at bottom of top photo B/-.C/-.D/- is black [or brown]  full colour, dense with O/  which we talk about below or red bottom photo B/-.cb/cb-.D/- is black  or brown sepia, dense becomes cream B/-.C/-.d/d  is black, (or brown) full colour(or sepia), dilute with O/ The reader should again note there are many variations on the theme of red. This is because there are Black reds, Brown reds, Sable reds, and Champagne reds,---AND all the Agouti versions of the same; on (as you will see ) a  Golden or Silver base. Creams have all the same number of subtle variations. Thus the judging process can not begin to differentiated and call them reds or creams only.

L short hair and long hair
Most laymen understand that a cat is either longhaired or short. Short hair is the dominant genetic trait but this is not quite so simple as is suggested, in our experience. None of our so-called longhaired cats have anything like the coat length of the Persian. They would probably be classified as semi longhairs if CCA had such a designation. Hair length seems to be a degree sort of thing. Hair discussions also need to talk about undercoat, its thickness lift or plushness if you will. We are not sure of those genetics but it now appears to us that this dense underfur is transmitted separately from long fur and that is a recessive. Our shorthairs seem to have some lift but not much after the second generation and we work to reduce this characteristic.  We now know there are 4 genes that deliver the "full puff" long hair.

       Mercury an agouti black, sepia,.dense   longhair not orange  -  actually we think he is the most cleaver of cats.

We see our LH Burmillas as having very much softer fur than the Chinchilla. The thickness of the undercoat is again somewhat variable but considerably less than the Persian cat resulting in a coat that lies down, and does not mat or require the large amount of care that Persians require. The other distinguishing feature is the LH Burmillas tends to look slightly out of focus or misty due to the fineness of the hair shafts. As with most LH cats they have the "wings" out to the side at the base of the ears. Neutering the cat often results in an increase in coat length.. Some of our kittens are born with pronounced long hair coat and we call them "woollies". Most of the kittens that become long hairs are difficult to distinguish as such until 8 weeks of age and sometime well beyond that. We hope to spend some time in the future on hair genetics, as it does play into color perception but we need a microscope to do that. Many long hair breeders probably could contribute much more to this discussion than we can at this time.

     Flamethrower a sepia red longhair

O orange or not orange 
This gene is sex linked and is a normal colour override. "The action of the O gene is to eliminate all eumelanistic pigmentation (black, chocolate, blue, etc) from the hair fibersso that all colour is red/cream.(see dilute above)
As it is sex linked it manifests differently between the sexes resulting in males being O/x (orange) or o/x (not orange) and females being OO (orange), O/o (tortie) and o/o (not orange). For more detail  on this see See Robinson’s Page 143 and 51).
 
As we have said above, "Red"  cats  can be  black or brown, full colour or sepia  At that time they are not colour differentiated in breeding standards, however the observer will be able to note variations.

Additionally  A/- cats can not be distinguished from the a/a cats.  All O cats show their underlying tabby markings.
 

Some variations on the theme of red. The dilute cream has less variations.

This photo of the full litter, with a very red boy. 

It is our observation the the colour red is "weeker" and more suseptable to the colour reducing genes i.e. cb and I and Wb (that are discussed later) . It is the torties that define these characteristics visually and judges need to understand this. The only true way to judge an O/o tortie is to have it placed beside an O/- cat, which is most often not practical.

An a/a.B/-.C/-.D/-.L/-.O/o female. Burmilla.
that is    non agouti, black, full colour, dense, shorthair, tortie
all those light patches are in fact red and that is on a cat that is not cb or I or Wb
 

Ta "abbytaby" markings or not
Robinson’s p 137 deals with all this and is very confusing. (Stephens and Robinson's do not agree. I have chosen Robinson's.) First para. re Abyssinian should be ignored at this time. Robinson’s are not clear what a ta/ta. Mc/- or mc/mc cat would look like.
Stephens says edit 1 p.13 last para. "The agouti (Abyssinian tabby or ticked tabby) appears to be dominant to other patterns..." She than photos the 3 patterns on page 14 they are 1) abbytaby, (my word); 2) Mackerel Tabby (bars down the side); 3) Classic tabby  (with the bulls eye on the side).
Robinson’s say:P 136 "It has been believed for many years that the three forms of tabby pattern were inherited as an allelic series; however, it now appears as if at least two, and possibly three, different loci are responsible for the various tabby patterns (Lorimer,1995). At one locus is the Abyssinian or ticked tabby pattern which is both epistatic to mackeral and to blotched tabby paterns." ... "proposed to use the letters Mc to represent the dominant Mackerel tabby pattern, while mc ... the recessive Classic, blotched tabby pattern."
We will henceforth remove their term "blotched "
Robinson’s then go on to say in their next para. and the reader will note that I have reordered these 2 quotes for better understanding (my view) and removed a misprint;
"Breeding data suggest a dominant modifying gene (...Sp) responsible for the spotted tabby phenotype" and "...two forms of spotted tabby pattern. The first is a modification of the mackerel in which the vertical stripes are discontinuous (giving) short bars or spots. Modification of the 'classic' tabby pattern ...for the round 'ocicat' type spots.
Stephens appears to concur on p 79 of edit 1
It thus falls out that a Ta/- will mask the Mc/- or mc/mc patterns, and all Sp/- cats.   Further, all mackerel and classic tabbies, are sp/sp.
I conclude and summarize as follows;
There are 3 separate gene locations operating on the tabby pattern. They are: Ta (abbytaby or not); Mc (mackerel tabby or classic); and Sp (spotting or not). If a cat is Ta/-, the Mc and the Sp can not be seen. It is epistatic.  If a cat is ta/ta (not) and sp/sp (not) a cat can be seen as Mc/- mackerel, or mc/mc  Classic. If a cat is ta/ta and Sp/- it will be spotted and if Mc/-, have full spots if mc/mc have "donut" spots.

again we have Mistick an  A/-.B/-.C/-.D/-.L/l o/o.Ta/- girl
This is Mistick daughter to Renaisance and Christian. and litter sister to Andromeda all shown above. It it this girl who proves that Christian carries I as we knew from his dame (see pedigree) and that he also carries wb - narrow band.

   This composite photograph shows another possible Burmilla variation.
This is the Spotted Tabby. A Black Spotted Golden Tabby F1 Burmilla girl, Tiverton's Andromeda. She is the daughter of Katie a Sable Traditional Burmese and Chicos Christian of Strathkirk, a Chinchilla Persian. We knew the boy was (I/i).  and) he is also Wb/wb. Andromeda belongs to Ms. Pat Slater of Tiverton Cattery

This is a photo of a very young, and fast moving, Silver Spotted Tabby F1 Burmilla kitten. He is the son of Ginette. a Platinum Traditional Burmese and Christian. Genetics then yields that both Ginette and Katie are  -/ta cats, and Christian may well be Sp but so too could the Burmese ladies. My bet is the ladies 'cause we have seen it so seldom but time will tell.

I inhibitor  - -this delivers a cat that is silver or ii (golden)
The silver inhibitor gene I, is separate from agouti gene and " operate only on the shredded or weakened phaeomelanin granules found on the agouti hairs (bands of yellow" -. (Stephens Edit #2 p21). We may well be wrong but we think of the basic hair as being manufactured in yellow or silver(no colour). The yellowing has varying degrees of rufousing {rufism is the spelling in Robinson’s](redness) considered to be a multiple gene effect (poly genetic)[rufus polygenes] Most of the cat family have this as seen in the Abyssinain and tigers. Even those rufoused cats often show very silver areas under the chin and on the belly.
Our gene pool is not rufoused and as such it can sometimes be difficult to determine if a cat is I/- or ii, particularly in the muted colours and or the colour reducing genes i.e. cb, I and Wb.

Just as we have found that the Burmese gene pool is not homogeneous for abytaby it also not for ii (golden)

This is Mistick's litter with Simon a Cream Burmese, and that boy in the front was a big surprise. This then has to mean that the silver boy got his stuff from Simon and that Simon is heterozygous (I/i) for this characteristic as the kittens sister is golden.
Simon is     a/a.B?.cb/cb.d/d.L/L.O/y.Ta/-.I/i.
the kitten is   A/a.B/-.cb/cb.D/d-.L/-.o/y.Ta/- I/i.

Burmese are I/I I/i and i/i.  We take that as proven. One of the curious aspects of Robinson’s and Stephens is that they do not spend much time discussing this recessive. All cats have the same genes, and like it or not they, the recessives, are doing something.

 When we first started thinking about this we thought we would have to do some breeding to prove our point. Then we remembered Angle is the sister of Simon and we had always wondered why her litter had not shown more golden kittens a we had known that Christian carried golden.

This photo of the full litter of Angel and Christian who are discussed above show what we understand to be a C/- ...o/x boy at the bottom left.
<There has been some recent (2006)discussion that 'I' is a different gene that 'i' and at a different location. This may well be true. That discussion takes the reader into the biochemistry of DNA and the "manufacturing process. Sufficient to say a not silver cat allows the golden to show. Perhaps the Abby breeders know more about this. >

This leads us to our conclusions on DM
We continue to have discomfort around the Dilute modifier -DM gene as outlined on page 141 of Robinson’s wherein they say "Dm is a dominant, that affects coat colour in dd (only) cats". They go on to suggest that all dilutes have two shades as follows with the second name in each case being paler:
blue and caramel
lilac and taupe
cream and apricot
We had long before recognized that there are two variations for each colour in our d/d "dilute" Burmese breedings, and Simon was particularly kept for his "powder" coat.
We think from what we see that there is a distinct possibility that DM is merely the manifestation of I/- and i/i in dilute cats. Using  Occam's razor, it is an elegantly simple explanation. It would be interesting to know P Turner's reasons for suggesting it. If you look at the diagrams Ms. Stephens presents (edit 1) on P 12 of edit #1 of the various colour dot deposition patterns not only do the full colour patterns show increasing space between the colour dots but the maltesed patterns show even large spaces.
What colour is the substrate on the i/i cat verses the I/- cat? It has got to affect what the eye sees. If so, all cats have this caramel/taupe/apricot colour variation, particularly the dilutes/maltesed cats.
Further to our point.

 

These are two composite photos of two of Pursephone's litter. Pursephone is a Burmilla who is
A/a.B/-.cb/cb.D/d.L/l.oo.Ta/-.I/-.Wb/- a, ticked cat  and. Zeus
a/a.b/b.cb/cb.D/d.L/l.o/o.Ta?.I/-. Wb/- sepia champagne smoked non-ticked cat was the sire.
Note the rich brown in a sable self boy above and the lighter brown sable sister to the right. What is of interest here is the light spot on this girl’s chest. She is (a/a.B/-.cb/cb) as is her brother.  There only 2 possible genes in play Wb and I that can define the difference.  Both cats are wb/wb (narrow band) and that will be discussed later but is not in "the game".
Both adults had Burmese parents and should be I/i. The kitten has to be I/- to have silver and we believe the light spot to be silver not white.  The only conclusion can be that she is a narrow band . We also suggest that the likely reason it shows be due to the sepia gene. This white chest patch is an issue on some Burmese. We postulate most and not all Burmese are golden, narrow band (. i/i. wb/wb). Those that show white patches are likely silver narrow band (I/- wb/wb). One breeding of a "locketed" Burmese to a golden tabby would prove the point. We therefore believe Robinson’s p151 on Lockets needs clarification.

We suggest this may be undesirable but should not be a disqualification factor in either the Burmese or Burmilla if we are correct.

We understand that the "Berrington" gene (and we hope we have spelled it correctly) may be related to this same I/- vs ii discussion. The  light Champagne Burmese is often said to have the ‘Berrington gene" and we think it may just be an I/- Burmese.

Recently we have heard that in CFA it is desirable that the Red Burmese show little of no baring ie tabby marking on the chest and legs. Given the above discussion and the reduced ability to differentiate the I/- Burmese from the i/i. It would seem to us that this might be a fundamental error and would select all Red Burmese to be I/- cats.
 

Wb Wide band and wb/wb "narrow band"
Robinson's p 142 in the section on the I gene says the following " The silver tabby exhibits a fairly low level of expression of this gene, while the chinchilla silver is a fine example of extreme phenotype...The smoke is the non-agouti form (a/a.I/-). The white undercoat is evident but each hair contains appreciably more pigment due to the lack of the additive inhibitory properties of the agouti factor."
In the prior para. Robinson's also says, "Because the melanin inhibitor gene is extremely variable in its expression, it can exhibit impenetrance, resulting in occasional cats with no visible white undercoat that nonetheless breed as smokes"
On Wide band Page 138 Robinson’s says Wide band (Wb), a hypothesized gene, is explained as "presumed effect of the gene is to widen the agouti band on the hairs" Robinson's p 138 and "Longhaired cats expose more of the agouti band and do not have such an obvious tabby pattern as do shorthairs." and "The golden tabby, shaded golden and chinchilla golden were developed from chinchilla longhairs.

Stephens agrees that Wb exists in her edit #2 in the chinchilla. (See page 21)
If Wb is dominant and if it exists in the chinchilla then it should be easy to track and observe. And it is.

  Again we give you Angel. This photo of the full litter of Angel and Christian who are discussed above

We know that Christian is  heterozygous Wb/wb because he produced

Mistick an  A/-.B/-.C/-.D/-.L/l o/o.Ta/- girl
This is Mistick daughter to Renaissance and Christian. and litter sister to Andromeda shown below. It is this girl who proves that Christian carries 'i' as we knew from his dame (see pedigree) and that he also carries wb - narrow band.

Thus we accept wb as fact, straightforward genetics and in the gene pool.
The reader should now note what exactly an agouti wide band cat looks like.

These photos are typical and you the reader have seen them before.
They are   o/o, and O/o and O/x cats.

1) They do not show the amount of tipping that is seen in the chinchilla, which leads us to our conclusion that there is a second gene that is recessive in the chinchilla. We will speak of this later and call it swb.
2) The boy is very tipped. What is the genetic difference if any?

First let’s define A/-  ... Wb/-  (wide band) and  wb/wb (narrow band cats) leaving O out of the mix

narrow band
  x    x

narrow band  cats
narrow band the girl above is blue and cb/ cb but does have the full necklace

 wide band
x
x 

swb super wide band

What we call Wb/-. swb/swb.   Both have sable paw pads and a brick nose
The reader will have noticed that the definition point between the Wb/- and the wb/wb is sometimes hard. We believe that to be due to the operation of swb, which delivers this 10-15 %, variability is the mix. Our observations at this time lead us to conclude that the swb is also straightforward genetic inheritance of a recessive.

Two smokes one probably a swb/swb , the other is not    Additionally one is B/- the other bb. Time will tell.

Second let’s talk of O
Robinson’s says on page 142 ‘"This gene appears to have a greater ability to suppress phaeomelanin pigment than eumelanin pigment
 .."

Three boys in a litter of A/- .. C/- kittens
Clearly the boy on the left is i/i; the boy on the right is either I/- or cb/cb.  Given that none of the seven kittens in the litter appeared to be cb/cb sepia we conclude him to be I/-. The reader should note then the difficulty of differentiating this in the red cat. The boy in the center with the smile illustrates that in the red cats Wb/- (wide band) is very clear. It shows as is a tipped cat with little colour and is instantly recognizable.
x
A narrow band and a wide band red. The reader my wonder how we know the girl on the left is narrow band. We didn't until she had a litter.
 

To prove the point this is a litter from the girl on the right  above with a red Burmese male

 
 

Smokes
 Smokes are wide band cats. The reader should know that we are on our own on this and in correspondence with Ms. Stephens, she does not agree, neither does she disagree. We present our case.
As we have said above it is the torties that often prove the case. Wide band cats in O have no (limited) colour (see above).
As can be seen below  the "silver" areas show no colour. The coloured areas show that cat is a smoke, no question. Thus smokes are wide band and will breed true.

Horizons Destiny
Born July 25 01 Sable Tortie Smoke female F3 to
Lucina (An F3 SH Sable Smoke female) and
Xanabu (a LH Red Tipped Silver (Cameo Shell) F3 male Burmilla) she is
a/a.B/-.cb/cb.D/-.l/l-.O/o.Ta/?.I/-.Wb/-

We could pile you up with examples, or you could go to the web site, but we will give you one more.

Strathkirk's Anastasia
(Stasia) was born Dec 31 1998. She is the daughter of Ch Calypso (a
Black Tipped Tortie SH Burmilla F1 female) & Not-So-Shy-
(a Sable Ticked Tabby LH Burmilla male).
She is a Black Tortie smoke female non-agouti,  black, full and dense colour, longhair, orange tortie,
abytaby, silver wideband cat  - can't be shaded - Black nose therefore she is a smoke
Genetics
a/a .B/b.C/-.D/-.l/l.O/o.Ta/-.I/-.Wb/wb
a/ a black nose a smoke a from calypso
Wb/wb see son noted below
B/b see son noted below

F3 Burmilla kittens born 23 June 02
Anastasia (LH Black Tortie Shaded Silver F2 Burmilla) X Viceroy
 (Cream male F+ European Burmese) F3 Burmilla
1 SH female cameo red?
2 SH male Sable self (one of these is a full expression champagne our first one)
 
 

There is one question.  All smokes we have seen are  I/- cats. Can a cat be i/i and Wb/-. as in a golden smoke? We have not seen one and we should have. Both Robinson’s and Stephens say there is such a cat.
Perhaps this is why Robinson’s says on p 142 "Because the melanin inhibitor gene is extremely variable in its expression, it can exhibit impenetrance, resulting in occasional cats with no visible white undercoat that nonetheless breed as smokes" but maybe not.

What are you, you beautiful boy!

We do believe that this exercise will force the fancy into defining it's colours based on genetics only, which should turn out to be an educational and worthwhile exercise, even it means undertaking some outcrossing in order to proove the genetics. 

 
This girl on the left is a young sable smoke female from Piper a Champagne Burmese and Pluvius a Sable Shaded Silver Burmilla. This litter did much to prove the genetics of Pluvius but is this girl who is interesting. Young smokes are often difficult to identify but as the reader  can see it is in the neck area that the smoking first becomes apparent.
Thus she is in all colour genetics similar to the Burmese with the one exception being that she is Wide band (Wb/-).

Conclusions
Genetics of colour and pattern
Below are two charts that integrate the information above. Hopefully they will help with an understanding of the varieties of Colors and Patterns in cats. They are in a form that would be familiar to cat breeders.
Patterns are separate from Colors but certainly play on what the eye sees, particularly when I, Wb, and swb, come into play (the polygenetic cop out). Torties and long hair are not considered for this exercise. 
The color terms used are generally those used in CCA; different fancies may use others (we have shown 2). These are the terms we will use, for now, when talking of "genetic color".
The reader will note there are effectively 11 colors. 12 if one can identify Russet Red.
Given 12 colors, 4 - 6 patterns, I, Wb,  swb, L, and O,   - that should give a rather large number of possibilities.
The reader will begin to understand why a number of Cat Fancies are slowly coming to the realization that cat coat can only truly be described by genetic terminology. For example how can one clearly describe the color Caramel? The reader will note the confusion left in the charts. Is it a golden (ticked) looking through blue/ indigo;  or is it a diluted brown sometimes called fawn?
These charts also provide a mathematical distribution of the possibilities -  - - the reader will note a number of errors and omissions in the standards that have been provided.


Below are two charts , one for colour and the lower one for pattern
As of feb'09 the reader should ignore the use of the terms to differentiate red  as in  russet red  and apricot  and the term caramel/fawn as note above. 
Frost is a colour suggested from Stephens and not currently used in the standard. It would be difficult to differentiate. Platinum = lilac