An ounce of blood is worth more than a pound of friendship.

Marcantonio Bassetti, Adorazione dei pastori, 1615.

In psychiatric classifications today, mood disorders are divided along the faultline of the presence or absence of manic or hypomanic periods in the patient’s personal history. The 1980 Diagnostic and Statistical Manual (DSM) classification recognized a disorder of depressed mood as a major depression. The presence of mania or hypomania, either in the examination or in the history, categorized the illness as a bipolar disorder. In the seesaw of psychiatric history, the mood disorders were seen either as two disorders or as a single entity. Defining the intimate connection between depression and mania is usually ascribed to the French authors Falret (1854) and Baillerger (1854), who described patients with both disorders at diVerent times in their illness. Karl Kahlbaum (1882) applied the term cyclothymia to a syndrome of circular disorders of excitement and depression that did not end in dementia. The concept was adopted by Kraepelin (1921) when he diVerentiated manic-depressive illness from that of dementia praecox. In Kraepelin’s view, the moods in manic-depressive illness included:

"certain slight and slightest colouring of mood, some of them periodic, some of them continuously morbid, which are to be regarded as the rudiment of more severe disorders . . . [or] pass without sharp boundary into personal predisposition . . . I have become more and more convinced that [manic-depressive] states represent manifestations of a single morbid process."

The groundwork for splitting manic-depressive illness into bipolar and unipolar disorders was laid by Leonhard (1995). In family studies he described patients suffering an endogenous illness with a history of mania who had a higher than expected prevalence of relatives with mania (labeled bipolar illness). These patients were distinguished from patients and their families with depressive illnesses only (monopolar illness).

The binary model was quickly supported by family history studies.It was incorporated in the DSM-III classification. This change in nosology encouraged the elaboration of new subtypes of bipolar disorder. Hospitalized depressed patients were divided into those with a history of mania and those who experienced hypomania into bipolar I and bipolar II groups. Unipolar depressed patients were divided into familial pure depressive disease, sporadic pure depressive disease, and depressive spectrum disease based on the family history of alcoholism and antisocial personality.

Additional variants were proposed in studies of less severely ill patients using criteria of family histories and statistical analyses of rating-scale items. The argument to classify unipolar and bipolar depression separately was challenged by Taylor and Abrams, first in literature reviews and then in a prospective clinical study of hospitalized, acutely ill patients with mood disorders and their first-degree relatives.

The family illness patterns were not clearly dichotomous. The most common mood disorder in the first-degree relatives of patients with bipolar disorder was recurrent depressive illness (unipolar disorder). The morbid risk (age-corrected prevalence) for unipolar disorder in the families of the bipolar patients was greater than the morbid risks for unipolar depression in the families of unipolar patients. The authors reviewed reports of mixed concordance for the two forms of mood disorder in twins, concluding that they could not distinguish the unipolar and bipolar patients by any family, psychopathological, treatment response or laboratory variable as long as depression was defined as melancholia.

They could not justify separating mood disorders by the presence of mania or hypomania, or by the clinical features of depression. In another detailed analysis, Goodwin and Jamison (1990) examined the course, epidemiology, family history, physiological measures, and similar criteria for the unipolar and bipolar forms of depression. These are assessed as either ‘‘less’’ or ‘‘more,’’ ‘‘younger’’ or ‘‘older,’’ ‘‘lower’’ or ‘‘higher’’ and other descriptive, nonquantitative differences.

Similar qualitative factors assess the incidence of the clinical signs of anxiety, anger, psychomotor agitation, mood lability, sleep time, pain sensitivity with the items rated as unipolar > bipolar or unipolar <>