PSR 0531+21 (Crab pulsar) is undoubtedly the best studied among
isolated pulsars, but several problems concerning the origin of the pulsed
radiation and the location of the emitting regions within the
magnetosphere are still open. To achieve more information on them we have
realized a computer accessible database containing an Atlas as complete
as possible of X and 
-ray pulse profiles appeared in the literature
since its discovery.
This Atlas could be not complete because some data may have not been
published yet. The Crab remnant and the pulsar, in fact, are frequently
observed by various astrophysical space instruments
(it is often used as a reference source) but not always these data
are presented in the literature. We ask then all the
scientists to help us in improving this database, which we hope can be
useful to many people,
by adding to the Atlas either the original data files of already published data
in order to avoid errors due to the figure digitization, or results which they
do not intend to publish elsewhere. Any other contribution to complete
the collection of the past observation of the Crab pulsar, and to continue
the data collection with new data, is welcome.
As a further comment, this Atlas is also useful to trace back the history
of the early X and 
-ray experiments which had often the Crab as the
primary target and which are now scarcely quoted in the literature, despite
some of them provided excellent data.
We used the Atlas profiles to investigate some subjects which have been
recently discussed in the recent literature. We do not find evidence for the
13.5 years periodicity of the P2/P1 ratio in the X and low energy
rays. We confirmed the time stability of this quantity in the
soft and medium X-ray range, while, at energies greater than about 30 keV,
some changes are apparent, but their occurrence and time scales are
still practically unknown. We obtained also the most complete picture of
the energy dependence of the P2/P1 and Ip/P1 ratios, which both
reach their maximum around 1 MeV. A further point studied here is the
shape of P1: a comparison of the Atlas profiles has shown
that it is not the same at all energies, but the right side is
characterized by an additional component in the hard X and low-energy 
rays. This result can be taken as an indication that the traditional
feature separation of the Crab pulse profile into P1, Ip and
P2 could not be appropriate. Conversely, one could adopt a different choice
separating the profile in an inner region, namely that between the
maxima of the peaks, and in an outer region. Such a model can be
physically based on a conical emission pattern having the vertex close to
one polar cap of the neutron star. The photons of inner region would then
be originated closer to the star magnetic pole than the others and therefore
have a different spectral distribution with a break at about
1 MeV. The overall stability of this pattern, in energy and time, is
indicated by the observed phase separation between the edges of the two
sections.
Acknowledgements
We are grateful to the FIGARO Collaboration for the permission to include in the Atlas some unpublished profiles. We are also indebted to W. Becker who gave us the data file of the ROSAT data. We are particularly grateful to M. Litterio for the help in organizing the WWW access to the computer database. This work has been carried out with the financial support of the Consiglio Nazionale delle Ricerche (GIFCO research contract).