Constantine Mifsud writes,
"What makes these single species so special that they are allocated to a
family in their own right? Are there any special ICZN rules regarding the
erection of families and sub-families, especially when single species are
concerned?"

No. The ICZN makes the rules and then steps out of the way. If you think
about it, what kind of rule could logically govern this case?

Here is an analogy with human genealogy. One family has many children and
is also closely related to other families. Another family has only one
child, but is not closely related to other families because of low numbers
of surviving children for several generations. In zoological taxonomy, we
try to place species that are literally most closely related together in
the same genera and families. In some cases, like Campanile, there is no
close living relative. Morphologically, the closest relatives are still not
very similar to Campanile. Molecular study would probably confirm this
distance, although I am not aware that this has been tried. In cases where
morphology and molecular analysis tell different stories, the molecular
analysis is usually more convincing, because shapes can change (think of
all the different breeds of dogs!) more easily than DNA and RNA.

But Constantine has a good point. Taxonomic ranks are NOT altogether
objective--not yet!--as can be demonstrated most easily by pointing to the
history of nomenclature in almost any given group of organisms. This brings
me to another taxonomic rule of thumb (whose origin I no longer remember).
On the average in the Animal Kingdom, there are about 3 named species per
genus. Some genera have hundreds of species, others have only one. But the
average is about 3. In my opinion, this is more likely to be a result of
the way that people think than a result of hard realities. If a genus has
many species, then it becomes evident that some can be grouped into
subgenera. There is no point to naming a new subgenus for every new
species, after all. Eventually, someone may decide that the subgenera
should be raised in rank to genera. Remember, whole families and orders
once started as genera like Nautilus, Ostrea, and Pecten as we all learn
more and approach the truth a little more closely. Resistance among
taxonomists to such changes in rank is least at the low ranks, greatest at
the high ranks. Until about 30-40 years ago, resistance to new phyla was
extremely high, and resistance to new kingdoms was total. Now we see
published schemes with 5, 7, and 9 kingdoms, all trying to show the real
distance among different groups a little better.

Andrew K. Rindsberg
Geological Survey of Alabama