oglobin is diluted and the amount in each blood-unit
greatly lessened.

_Pernicious anaemia_ is a rarer disease than chlorosis, occurs usually
later in life, and is distributed nearly equally between the two sexes.
But it is of great importance because of its almost uniformly fatal
termination, though its downward course is generally broken by
temporary improvement on one or more occasions. The symptoms are those
of a progressive anaemia, in which gastro-intestinal disturbance usually
plays a large part, and nervous symptoms are common, and they become at
last much more severe than those of any secondary anaemia. The patient
may die in the first attack, but more usually, when things seem to be at
their worst, improvement sets in, either spontaneously or as the result
of treatment, and the patient slowly regains apparent health. This
remission may be followed by a relapse, that again by a remission, and
so on, but as a rule the disease is fatal within, at the outside, two or
three years.

The prime cause of the disease is not known. It seems probable indeed
that the causal factors are numerous. Severe malarial infection,
syphilis, pregnancy, chronic gastro-intestinal disease, chronic
gas-poisoning, are all, in different cases, known to have been causally
associated with it, and it is probable that a congenital weakness of the
bone-marrow has often to do with its production, as in many cases a
family or hereditary history of the disease can be obtained. The
condition is now regarded as a chronic toxaemia, partly because of the
clinical symptoms and pathological appearances, partly because analogous
conditions can be produced experimentally by such poisons as saponin and
toluylendiamin, and partly because of the facts of _bothriocephalus_
anaemia. The site of production of the toxin, or toxins, for it is
possible that several may have the same effect on the blood, is possibly
not always the same, but must often be the alimentary canal, as
_bothriocephalus_ anaemia proves. Not all persons affected with this
intestinal tapeworm contract the disease, but only those in whose
intestines the worm is dead and decomposing or sometimes only "sick."
The expulsion of the worm puts an end to the absorption of the toxin and
the patients recover. No adequate explanation of the formation of the
toxin in the immense majority of the cases, in which there is no
tapeworm, has yet been given. It is certain that no organism as yet
known is c