The preceding quotations, taken directly from the most recent edition of textbooks used often in dermatology and dermatopathology, reveal how vast is confusion in regard to the designations follicular mucinosis, alopecia mucinosa, mycosis fungoides, and cutaneous T-cell lymphoma, not only concerning definition and essential character of each of them, but of relationships among them. We seek in this work to illuminate the entire matter by first framing it in historical perspective, then setting forth our own observations pertinent to clinical, histopathologic, and biologic aspects of it, and, last, proposing a concept that synthesizes all that is known about the subject in a way that enlightens at the same time that it dispels bewilderment. Early in the course of this endeavor, we will explain how the terms follicular mucinosis, alopecia mucinosa, and mycosis fungoides came to be and how they are employed currently in publications of all kinds, chiefly in articles and books. In that way, a reader is alerted, from near the outset, to pitfalls in terminology that impede comprehension of the issues under discussion here. In order to achieve that desideratum, it is requisite that we address briefly considerations that pertain both to embryologic development and to histologic attributes of the skin, and especially of the folliculosebaceous-apocrine unit, that structure being the locus of "follicular mucinosis," a finding histopathologically that is essential to diagnosis of "alopecia mucinosa." Parenthetically, as will be learned as the reader reads on, eccrine units are affected episodically in the pathologic process that engages us here, but not by mucinosis.

All epithelial structures of skin are derived from surface ectoderm. At about 10 weeks in the life of an embryo, two primorda (anlagen) appear on the undersurface of that all-enveloping primitive epithelium: one gives rise to the folliculosebaceous-apocrine unit and the other to the eccrine unit. In the former unit, germinative cells in crescentic-shaped aggregations, known for purpose of succintness as follicular germs (Fig. 1 ), proliferate and give rise to the entire future hair follicle, as well as to derivatives of it, namely, the sebaceous gland and duct, and the apocrine gland and duct. Contiguous with a follicular germ and seated immediately beneath it is a follicular papilla, to wit, a cluster of mesenchymal cells that are in contiguity with similar appearing cells and in continuity with fibrous tissue that will make up the future perifollicular sheath. In the sixteenth week of embryonic life, epithelial cells become crowded at three distinct loci on one side of a developing infundibulofollicular structure (the infundibulum being a funnel-shaped part of the epidermis, in contrast to the more bar-like appearance of the surface portion of the epidermis), each of those protuberances being called bulges. The lowest bulge from the follicle itself becomes the site of attachment for smooth muscle fascicles of hair errection, the middle bulge from the junction of infundibulum and follicle develops into a sebaceous unit, and the uppermost bulge from the infundibulum eventuates in an apocrine unit.¹³

By the time that bulges appear along the side of an evolving infundibulofollicular structure, the still differentiating follicle is seen to consist of three distinct parts (Fig. 2 ), to wit, the isthmus, which is delimited above by the entry of the sebaceous duct into the base of infundibular epidermis and below by desquamation of cornified cells of the inner sheath, the stem, which is bounded above by the site at which cornified cells of the inner sheath desquamate and below by Adamson's fringe, that is, the place where the future hair shaft becomes anucleate, and the bulb, which is demarcated by Adamson's fringe above and the base of the follicle below.¹⁴ The bulb of a mature follicle in anagen consists mainly of matrical cells, those situated in the center of the bulb maturing to become corneocytes of hair shaft and those adjacent to it becoming corneocytes of inner sheath. Matrical cells at the periphery of the bulb become cells of the outer sheath, so named because that sheath envelops the inner sheath.

It must be emphasized that the infundibulum, which is in continuity with both surface epidermis and the uppermost (isthmic) portion of a follicle, is wholly epidermal, not follicular.¹⁴ In short, a follicle can be separated arbitrarily into three parts (bulb, stem, and isthmus) and the epidermis can be divided equally arbitrarily into two parts (infundibulum and surface).

Eccrine units come into being as a consequence of proliferation of germinative cells arrayed in nubben-shaped aggregations independent of those that eventuate in formation of the folliculosebaceous-apocrine unit. The eccrine germ, like the folliculosebaceous-apocrine germ, derives from surface ectoderm, but, unlike the folliculosebaceous-apocrine germ, it is unaffiliated with a papilla.

The designation "alopecia mucinosa" was introduced in 1957 by Herman Pinkus, who by it sought to describe a combination of findings in a condition he believed to be distinctive and unrecorded previously.¹⁵ According to Pinkus, the condition was typified clinically by discrete follicular papules within plaques that were alopecic. On examination histopathologically, mucin was seen to be housed in infundibular, follicular, and sebaceous epithelium (Pinkus himself believed that the infundibulum was a component of a follicle). As of this writing in 2003, dermatologists and pathologists affirm, nearly universally, that alopecia mucinosa manifests itself in two very different ways, one being associated with "cutaneous T-cell lymphoma," mycosis fungoides in particular, and the other being an inflammatory process that either is unrelated entirely to mycosis fungoides or that transforms uncommonly into mycosis fungoides. The terms "primary" and "secondary," as well as "idiopathic" and "symptomatic," have been used to categorize alopecia mucinosa, the idea for those characterizations being that examples of alopecia mucinosa which seem to develop de novo in the absence of lymphoma are "primary" or "idiopathic," whereas those that represent a peculiar manifestation of an already established "cutaneous T-cell lymphoma" are "secondary" or "symptomatic."

The term "follicular mucinosis" was proposed in 1959 by Jablonska and coworkers as a synonym for what Pinkus had called alopecia mucinosa, and that is just how it is used in most quarters today.¹⁶ Over the years, a few authors have remarked that "follicular mucinosis" should be employed only as a description of a particular finding histopathologically, in contrast to "alopecia mucinosa" which should be used to name a disease that is identifiable clinically and histopathologically, the sine qua non for the latter malady being the finding by conventional microscopy of follicular mucinosis. For example, Pinkus¹⁷ and Hyman,¹⁸ both in 1962 and independent of one another, opined that follicular mucinosis should not be conceived of or utilized as a synonym for alopecia mucinosa. Pinkus stated crisply that "follicular mucinosis designates the histologic picture that may be found as the primary change in alopecia mucinosa or as a symptomatic feature in certain cases of mycosis fungoides" and Hyman said tersely that "this [follicular mucinosis] is a reaction pattern rather than a specific disease." It was Ackerman, in collaboration with Hempstead, who, in 1985, proposed that follicular mucinosis is a pattern of epithelium characterized by mucin situated mostly in infundibula and sebaceous glands, and that is analogous, therefore, to other distinctive histologic patterns of cutaneous epithelium, the most common of those being epidermolytic hyperkeratosis, focal acantholytic dyskeratosis, cornoid lamellation, and pale-cell acanthosis.¹⁹

Unfortunately, the distinctive pattern of epithelium under discussion here is not captured accurately by the term "follicular mucinosis," the reason being that sections of tissue cut routinely and that house those findings nearly always are compromised by being oriented somewhat obliquely. That being the case, mucin is much less apparent in true follicular epithelium than in infundibular epidermis and in sebaceous lobules. Therefore, the designation "follicular mucinosis" is misleading; more often than not, no mucin is identifiable in follicular epithelium in a particular section of tissue that sports attributes of "follicular mucinosis," the dominant finding in such a section actually being infundibular mucinosis. Moreover, whatever the merit or lack of it, the appellation "follicular mucinosis" should not be used to refer to a specific disease, but only to particular histopathologic findings.

Because alopecia mucinosa has been claimed to be associated with or to transform into mycosis fungoides and because changes of follicular mucinosis are encountered episodically in sections of biopsy specimens of mycosis fungoides (and sometimes in other manifestations of "cutaneous T-cell lymphoma"), it is worthwhile to consider how the terms mycosis fungoides and cutaneous T-cell lymphoma (CTCL) came into common parlance.

Alibert was the first, in 1806, to write about a disease he named, redundantly, mycosis fungoides.²⁰ He chose that name because the condition presented itself as mushroom-shaped tumors that he believed resulted in death invariably, and in a relatively short period of time. In the 125 years that ensued, mycosis fungoides was considered by dermatologists everywhere to be fatal inevitably. Only in the 1940s was the notion that flat lesions of mycosis fungoides could be diagnosed clinically, with repeatability and reliability, that concept having been spawned by Keil²¹ and championed, not only as a tenet clinically but as a principle histopathologically, by Ackerman, commencing in the 1970s.²² Until that time, findings in sections of tissue of biopsy specimens taken from patches of mycosis fungoides were said to be non-specific and non-diagnostic, a contention even today of many dermatopathologists and of other students of mycosis fungoides.

In 1902, Brocq described a group of diseases that reminded him vaguely of psoriasis, thereby prompting him to call it parapsoriais. In his classification of parapsoriasis, Brocq distinguished three groups that he deemed to be related to one another.²³ What follows now is more than a tad dizzying, but it illustrates poignantly how exceedingly difficult it is, and has been, for a serious student of this subject to come to grips with the matter in a meaningful way. Brocq gave the name "parapsoriasis lichenoides" to parakeratosis variegata²⁴ and he called exanthema psoriasiforme lichenoides²⁵ "parapsoriasis en goutte." The third in this "family" of diseases he termed "parapsoriasis en plaque," a condition that he, himself, in 1897, had described first by the phrase "erythrodermies pityriasique en plaque."²⁶ Brocq had come to appreciate that what he referred to as parapsoriasis could present itself clinically in a manner indistinguishable from that of early mycosis fungoides,²⁷ but, despite the insight, he nevertheless regarded all diseases gathered by him under the rubric "parapsoriasis" as something other than mycosis fungoides. For most of the 20th century, Brocq's ideas were embraced fervently by dermatologists everywhere.^28–33 At long last, and only in the last two decades of that century, colleagues worldwide came to acknowledge that "large plaque parapsoriasis" is mycosis fungoides and not a "precursor" (premycotic or eczematous stage) of that lymphoma.^34–39 Nearly all dermatologists, however, argue (in our view wrongly) to this day that "small plaque parapsoriasis" is not mycosis fungoides.

Other manifestations of mycosis fungoides akin to Brocq's "large plaque parapsoriasis" also were acknowledged to exist, but because they were not recognized for what they were, namely, flattish lesions of mycosis fungoides, they were given curious appellations, among those being xantherythroderma perstans, pokiloderma vasculare atrophicans, pagetoid reticulosis (Woringer-Kolopp disease), Sézary's syndrome, digitate dermatosis, and guttate parapsoriasis/dermatosis. All those designations reflect an effort to capture attributes morphologically as they present themselves clinically mostly, e.g., erythroderma in the case of Sézary's syndrome, atrophy of poikiloderma vasculare atrophicans, the shape of fingers pressed against the skin of digitate dermatosis, and drop size and shape of lesions of guttate dermatosis, but they convey little about findings histopathologically and nothing about the fundamental nature of the pathological process that truly is unifying of them, namely, mycosis fungoides.

Of the "cutaneous T-cell lymphomas," mycosis fungoides is the most common by far, that all embracing phrase, proposed by Edelson in 1975, having come to be used synonymously with mycosis fungoides by almost everyone, even though, in actuality, it is generic for a group of cutaneous lymphomas of T-cell type.⁴⁰ Other lymphomas that qualify as being "cutaneous T-cell" are lymphomatoid papulosis, anaplastic large cell lymphoma, adult T-cell lymphoma/leukemia, and, possibly, some examples of Hodkin's disease.⁴⁰ Edelson, by introducing of the designation and concept "cutaneous T-cell lymphoma," sought to convey that all of the lymphomas just mentioned are related to one another by having in common "primary skin infiltration by malignant T cells." As an aside it should be mentioned that, in times past, all the cutaneous lymphomas alluded to by us thus far were termed "reticuloses," some of them being called "eosinophilic reticulosis" and others "eosinophilic granuloma."

Mycosis fungoides at first presents itself clinically as macules and then as patches; macules may become papules and patches may become plaques. Later still, and uncommonly, those lesions may develop into nodules and tumors. In fact, most patients with mycosis fungoides have only patches and/or slightly raised plaques for a lifetime with no progression of those lesions to tumors. When flat or very subtly elevated lesions are examined histopathologically, lymphocytes are seen to be present around venules of the superficial plexus, in variable numbers in dermal papillae, and as solitary units positioned in the epidermal basal layer and spinous zone, the latter in company with scant spongiosis. Often the lesion is topped by thin mounds of parakeratosis joined by a tad of plasma. Well defined plaques are typified by more dense infiltrates of lymphocytes around venules, in patchy lichenoid fashion in conjunction with wiry bundles of collagen arrayed haphazardly, and in the basal and spinous layers of the epidermis as solitary units and, later, in the spinous (and sometimes granular and cornified) layer as collections of different sizes, known wrongly as Pautrier's microabscesses (it was Darier, not Pautrier, who first described them,⁴¹ and they are not abscesses, composed as they are of lymphocytes, rather than neutrophils). Nodules and tumors of mycosis fungoides (except for pendulous folds named misleadingly "granulomatous slack skin" and typified histopathologically by patchy perivascular infiltrates of lymphocytes and by striking granulomatous inflammation throughout the dermis and far into the subcutaneous fat) are formed by nodular and diffuse infiltrates of atypical lymphocytes distributed throughout the dermis and often, too, in the subcutaneous fat. At that late stage, infiltrates of abnormal lymphocytes in the skin and subcutaneous fat are joined nearly always by similar infiltrates that have become overt in internal organs. It is reasonable to assume that the process known as mycosis fungoides is systemic from the outset and simply manifests itself first in the skin.

Sézary's syndrome is a widespread erythrodermic manifestation of mycosis fungoides, an expression of it that was called attention to originally in 1892 by Hallopeau and Besnier⁴² and again in 1949 by Sézary, for whom the condition is named eponymically by virtue of his having detected atypical mononuclear cells in the circulation of persons affected.⁴³

"Follicular mycosis fungoides," a term fast gaining popularity, is a presentation of mycosis fungoides that affects infundibular epidermis predominantly, with little or no involvement of surface epidermis. In reality, most examples of "follicular mycosis fungoides" are "infundibular mycosis fungoides," the reason being that the infundibulum, not the follicle, is the major locus of the infiltrates of abnormal lymphocytes. For some authors, the finding of mucin in infundibular epithelium is a criterion for diagnosis of follicular mycosis fungoides, whereas for other commentators on the matter, the finding of follicular mucinosis in mycosis fungoides excludes from consideration a diagnosis of follicular mycosis fungoides.

It merits mention that the finding histopathologically of follicular mucinosis has been reported on in almost all of the different manifestations of "cutaneous T-cell lymphoma," among those being conventional mycosis fungoides, follicular mycosis fungoides, Sezary syndrome, syringotropic lymphoma, and Hodgkin's disease.