We present the 15‐μm extragalactic source counts from the Final Analysis Catalogue of the European Large Area ISO Survey (ELAIS) Southern hemisphere field S1, extracted using the Lari method. The large number of extragalactic sources (∼350) detected over this area between about 0.5 and 100 mJy guarantee a high statistical significance of the source counts in the previously poorly covered flux density range between IRAS and the Deep ISOCAM Surveys. The bright counts in S1 (S15 μm≳ 2 mJy) are significantly lower than other published ISOCAM counts in the same flux range and are consistent with a flat, Euclidean slope, suggesting the dominance of a non‐evolving population. In contrast, at fainter fluxes (S15 μm≲ 2 mJy) our counts show a strong departure from no‐evolution models, with a very steep super‐Euclidean slope down to our flux limit (∼0.5 mJy). Strong luminosity and density evolutions of the order of, respectively, L∝ (1 +z)3.0 and ρ∝ (1 +z)3.5 are needed at least for the population of star‐forming galaxies in order to fit the counts and the redshift distributions observed at different fluxes. A luminosity break around 1010.8L⊙ must be introduced in the local luminosity function of starburst galaxies in order to reproduce our sharp increase of the counts below 2 mJy and the redshift distributions observed for 15‐μm sources at different flux levels. The contribution of the strongly evolving starburst population (down to 50 μJy) to the 15‐μm cosmic background is estimated to be ∼2.2 nW m−2 sr−1, which corresponds to ∼67 per cent of the total mid‐infrared background estimate.

We present the final band‐merged European Large‐Area ISO Survey (ELAIS) Catalogue at 6.7, 15, 90 and 175 μm, and the associated data at U, g′, r′, i′, Z, J, H, K and 20 cm. The origin of the survey, infrared and radio observations, data‐reduction and optical identifications are briefly reviewed, and a summary of the area covered and the completeness limit for each infrared band is given. A detailed discussion of the band‐merging and optical association strategy is given. The total Catalogue consists of 3762 sources. 23 per cent of the 15‐μm sources and 75 per cent of the 6.7‐μm sources are stars. For extragalactic sources observed in three or more infrared bands, colour–colour diagrams are presented and discussed in terms of the contributing infrared populations. Spectral energy distributions (SEDs) are shown for selected sources and compared with cirrus, M82 and Arp220 starburst, and active galactic nuclei (AGN) dust torus models. Spectroscopic redshifts are tabulated, where available. For the N1 and N2 areas, the Isaac Newton Telescope ugriz Wide Field Survey permits photometric redshifts to be estimated for galaxies and quasars. These agree well with the spectroscopic redshifts, within the uncertainty of the photometric method [∼10 per cent in (1 +z) for galaxies]. The redshift distribution is given for selected ELAIS bands and colour–redshift diagrams are discussed. There is a high proportion of ultraluminous infrared galaxies (log10 of 1–1000 μm luminosity Lir > 12.22) in the ELAIS Catalogue (14 per cent of 15‐μm galaxies with known z), many with Arp220‐like SEDs. 10 per cent of the 15‐μm sources are genuine optically blank fields to r′= 24: these must have very high infrared‐to‐optical ratios and probably have z > 0.6, so are high‐luminosity dusty starbursts or Type 2 AGN. Nine hyperluminous infrared galaxies (Lir > 13.22) and nine extremely red objects (EROs) (r−K > 6) are found in the survey. The latter are interpreted as ultraluminous dusty infrared galaxies at z∼ 1. The large numbers of ultraluminous galaxies imply very strong evolution in the star formation rate between z= 0 and 1. There is also a surprisingly large population of luminous (Lir > 11.5), cool (cirrus‐type SEDs) galaxies, with Lir−Lopt > 0, implying AV > 1.

Aims. We study the evolution of the luminosity function (LF) of type-1 and type-2 Active Galactic Nuclei (AGN) in the mid-infrared, derive the contribution of the AGN to the Cosmic InfraRed Background (CIRB) and the expected source counts to be observed by Spitzer at 24 µm. Methods. We used a sample of type-1 and type-2 AGN selected at 15 µm (ISO) and 12 µm (IRAS), and classified on the basis of their optical spectra. Local spectral templates of type-1 and type-2 AGN have been used to derive the intrinsic 15 µm luminosities. We adopted an evolving smooth two-power law shape of the LF, whose parameters have been derived using an un-binned maximum likelihood method. Results. We find that the LF of type-1 AGN is compatible with a pure luminosity evolution. A small flattening of the faint (L 15 < L * 15 ) slope of the LF with increasing redshift is favoured by the data. A similar evolutionary scenario is found for the type-2 population with a rate k L ranging from ∼1.8 to 2.6, depending significantly on the adopted mid-infrared spectral energy distribution. For type-2 AGN a flattening of the LF with increasing redshift is suggested by the data, possibly caused by the loss of a fraction of type-2 AGN hidden within the optically classified starburst and normal galaxies. The type-1 AGN contribution to the CIRB at 15 µm is (4.2-12.1) × 10 −11 W m −2 sr −1 , while the type-2 AGN contribution is (5.5-11.0) × 10 −11 W m −2 sr −1 . We expect that Spitzer will observe, down to a flux limit of S 24 µm = 0.01 mJy, a density of ∼1200 deg −2 type-1 and ∼1000 deg −2 type-2 optically classified AGN. Conclusions. AGN evolve in the mid-infrared with a rate similar to the ones found in the optical and X-rays bands. The derived total contribution of the AGN to the CIRB (4-10%) and Spitzer counts should be considered as lower limits, because of a possible loss of type-2 sources caused by the optical classification.

We present the first determination of the 15 m luminosity function of galaxies from the European Large Area ISO Survey (ELAIS) southern fields. We have adopted a new criterion to separate the quiescent, nonevolving and the starburst, evolving populations based on the ratio of mid-infrared to optical luminosity. Strong evolution is suggested by our data for the starburst galaxy population, while normal spiral galaxies are consistent with no evolution. The starburst population must evolve both in luminosity and in density with rates of the order L(z) / (1 þ z) 3:5 and (z) / (1 þ z) 3:8 , respectively, up to z $ 1. The evolutionary parameters of our model have been tested by comparing the model predictions with other observables, like source counts at all flux density levels (from 0.1 to 300 mJy) and redshift distributions and luminosity functions at high z (0:7 < z < 1:0 from Hubble Deep Field North [HDF-N] data). The agreement between our model predictions and the observed data is remarkably good. We use our data to estimate the star formation density of the universe up to z ¼ 0:4, and we use the luminosity function model to predict the trend of the star formation history up to z ¼ 1.

The comparison between the new Spitzer data at 24 mm and the previous ISOCAM data at 15 mm is a key tool to understand galaxy properties and evolution in the infrared and to interpret the observed number counts, since the combination of Spitzer with the Infrared Space Observatory cosmological surveys provides for the first time the direct view of the universe in the infrared up to . We present the prediction in the Spitzer 24 mm z տ 2 band of a phenomenological model for galaxy evolution derived from the 15 mm data. Without any "a posteriori" update, the model predictions seem to agree well with the recently published 24 mm extragalactic source counts, suggesting that the peak in the 24 mm counts is dominated by "starburst" galaxies like those detected by ISOCAM at 15 mm but at higher redshifts ( instead of ). 1 Շ z Շ 2 0 . 5Շ z Շ 1.5