They are all the Holotype.  Just because part of a specimen is removed does not make it "unpart" of the specimen.  If a museum gives separate catalog numbers to them that is an artifact of the museums catalog system.

It is a relatively common practice to dissect types as part of the description process and the parts are still part of the the holotype.

Now if the parts get mixed up that is also covered by Article 73

"If a subsequent author finds that a holotype which consists of a set of components (e.g. disarticulated body parts) is not derived from an individual animal, the extraneous components may, by appropriate citation, be excluded from the holotype (material may be excluded from a hapantotype if it is found to contain components representing more than one taxon."

Agreed. The holotype is the complete specimen, i.e. the sum of all of its parts, no matter whether these parts are deposited in one or more museum collections.

Unfortunately, some museum numbering systems do not preview a situation like this. In my opinion, the same number shoud be given to all parts, as I use to do with mammals, for example: dry round skin prapared, body with organs (without skin), or skeleton, all receive the same number. Although it may seem weird to some (because one of these collections makes a non sequencial numbering), this method insures no confusion among other million specimens. All parts MUST be consider as one only specimen. 

Paulo Auricchio

 Thanks for the answers, guys. I figured it was, as Will Reeves said, just an artifact of different numbering schemes for different parts of a collection, which could be fixed by giving parts of specimens the same number. 

Agreed, when  rock containing compressed plant fossils is split during collection in the field or in the laboratory  the fossil presents as two complementary halves, known as Part and Counterpart.  These are often numbered separately in museum collections  but are  the same specimen and should be numbered as one number - though no problem using 'a' and 'b' for the two portions.   Many fossil and living plants are first described from small pieces - for example pinnae of the large fern Angiopteris - the holotype is the name permanently attached to these, and, yes, the name of course applies to the whole plant, when more of it may be collected.

This reply arises partly out of discussions with Prof Zhou Zhiyan at the Nanjing Institute of Geology and Palaeontology, China, an dwith Cedric Shute at the British Natural History Museum in London.

I don't like decimal suffixes to catalogue numbers, but this is the sort of situation where they are very useful. And yes, all the parts are the holotype.

in your case they are all holotype. Numbering depends of Museum.

in case of colonial animals - like cnidarians and sponges - perhaps some bryozoans as well - there is a case when pieces of the same colony deposited in several museums. In this case we use 'schizotype'. But definitely it is not your case

Agree with all hat was written above. They are all holotype  because belong to the same specimen. I think that the numbers do not matter. It would haave been different if the species were described only on the bases of the slin, for example, and later you or some one other completed the description with other part of the body from other specimens of the same specie.

Describing a new species, you should explicitly start with the description of the complete specimen, figure it, designate it explicitly as the holotype, and cite the museum collection where it will be deposited without numbers. That would fulfill everything of Article 16.4 of the ICZN.

Then you should describe and figure characteristic features based on prepared parts of the holotype and give the numbers of the museum collections which in your case are differing. 

They are all parts of a single holotype. we have had a couple of examples where bivalve shells are at one museum and the body of the same animal is in another museum. both are parts of the same animal and remain the holotype,  just a case of curatorial confusion.

Another museum curator here. I agree, all parts are still parts of the holotype. Numbering is irrelevant. Labels should clearly indicate the status of the part.

In botany the fragments  are schizotypes of same  Holotype- Isnt it????

It's not quite as straightforward as the respondents so far suggest. The key aspect of a holotype is: What did the describer examine when the description was prepared? If a skull was examined and a skin subsequently found that others associate with the skull (or vice versa), then caution should be exercised when drawing taxonomic conclusions from any part of the body other than that that was originally examined. This is especially relevant to fossil material where association of parts to form one individual may be controversial. Only the parts that the describer assigned to the individual should strictly-speaking be considered as part of the holotype.

Of course, none of Science is straightforward.  As  Karl Popper and his students indicated it proceeds on the basis of hypothesis creation and testing, like driving piles ever deeper into a swamp as a foundation of a house but never quite reaching the rock bottom.  Taxonomic naming however is not a Science as such, so the name normally will always be associated with the bit that it was originally given to, which has no theoretical significance and is just operational, i.e. has no wider meaning than a name.  

I don't think the term schizotype exists in zoological nomenclature. And my previous response was in relation to the actual question, a planned or ongoing description, not old material. Treatment of old museum material requires a lot of detective work and careful documentation to guarantee traceability but I don't think that was the question here.

Christopher is correct. Species names are "tags" associated with a single specimen (or type series) or part of a specimen. That tag should not be confused with a species hypothesis that may be based on additional specimens or parts of specimens (if the holotype specimen comprised, say, a single element of a multipart skeleton). The holotype plus parts thought (but not proven) to be belong to the same individual comprises a "Hypodigm" (or reconstructed individual) as used by some vertebrate palaeontologists (following Simpson's usage). Nevertheless, the holotype will forever remain the single skeletal element originally designated.

Regarding museum numbering systems, there are at least two contradicting needs. 1. each object needs to be identified beyond doubt, 2. parts of the same individual need to be identified as such. Number 1 can be accommodated by assigning each object a unique number, but then we can't be sure which parts belong to the same individual. Number 2 can be accommodated by assigning all parts of an individual the same number but then we can't be sure we have all objects, which is particularly important with loans. What we need is two numbers, one for the individual, common to all parts, another for each object, unique to each one. Instead of numbers, a text comment works too. Few museum collections are consistent in applying the same numbering system over time though.

To me also these are all parts of the same type specimens (independent of the kind of type it is)

I think in cases like this a numbering system using decimals (eg. MB.R.2181.3) or (still OK but less versatile in my eyes) letters (Ce 108a) for different parts of the same organism (or any other unit of the collection in the case off several specimens on the same unit (like foraminifers mounted on a slide, several fossils on a single plate etc.) is the best solution.

In botany, all parts of a holotype must only belong to a single gathering. It does not matterr whether they're split into separate sheets of herbarium. The only demand is to mark such parts (on their labels) to show they belong to a single gathering (a sample taken at the same time and from the same place and population).

what difference between schizotype and isotype (particularly in this case) ?

in fact the term 'schizotype' (schizoholotype or schizoparatype) does not really exist in zoology and not regulated by ICZN but it is a way to explain that the specimen was split and there are few cases when it can be attributed. definitely it is not like skin and bones or shell and soft body

Let's imagine a scenario even more "troubled", where more than two museums are involved. I think that can be cited for example: Holotype: ANSP [number], shell; MZSP [number], soft parts; USNM [number], radula. They all are the holotype, the same specimen.

Dear Nathaniel,

I agree with Dr. Reeves (article 73) and also with the "problematic scenario" proposed by Dr. Cunha. Both pieces of this specimen remain a holotype.

All the best.