High-quality, blue-violet spectroscopic data are collected for 24 stars that have been classified as type O3 and that display the hallmark N iv and N v lines. A new member of the class is presented; it is the second known in the Cyg OB2 association, and only the second in the northern hemisphere. New digital data are also presented for several of the other stars. Although the data are inhomogeneous, the uniform plots by subcategory reveal some interesting new relationships. Several issues concerning the classification of the hottest O-type spectra are discussed, and new digital data are presented for the five original O3 dwarfs in the Carina Nebula, in which the N iv, N v features are very weak or absent. New spectral types O2 and O3.5 are introduced here as steps toward resolving these issues. The relationship between the derived absolute visual magnitudes and the spectroscopic luminosity classes of the O2-O3 stars shows more scatter than at later O types, at least partly because some overluminous dwarfs are unresolved multiple systems, and some close binary systems of relatively low luminosity and mass emulate O3 supergiant spectra. However, it also appears that the behavior of He ii 4686, the primary luminosity criterion at later O types, responds to other phenomena in addition to luminosity at spectral types O2-O3. There is evidence that these spectral types may correspond to an immediate pre-WN phase, with a correspondingly large range of luminosities and masses. A complete census of spectra classified into the original O3 subcategories considered here (not including intermediate O3/WN types or O3 dwarfs without N iv, N v features) totals 45 stars; 34 of them belong to the Large Magellanic Cloud and 20 of the latter to 30 Doradus.

7We report the orbital distribution of the trans-neptunian objects (TNOs) discovered during the Canada-France Ecliptic Plane Survey (CFEPS), whose discovery phase ran from early 2003 until early 2007. The follow-up observations started just after the first discoveries and extended until late 2009. We obtained characterized observations of 321 sq.deg. of sky to depths in the range g ∼23.5 -24.4 AB mag. We provide a database of 169 TNOs with high-precision dynamical classification and known discovery efficiency. Using this database, we find that the classical belt is a complex region with sub-structures that go beyond the usual splitting of inner (interior to 3:2 mean-motion resonance [MMR]), main (between 3:2 and 2:1 MMR), and outer (exterior to 2:1 MMR). The main classical belt (a=40-47 AU) needs to be modeled with 1 Based on observations obtained with MegaPrime/MegaCam, a joint project of CFHT and CEA/DAPNIA, at the Canada-France-Hawaii Telescope (CFHT) which is operated by the at least three components: the 'hot' component with a wide inclination distribution and two 'cold' components (stirred and kernel) with much narrower inclination distributions. The hot component must have a significantly shallower absolute magnitude (H g ) distribution than the other two components. With 95% confidence, there are 8000 +1800 −1600 objects in the main belt with H g ≤ 8.0, of which 50% are from the hot component, 40% from the stirred component and 10% from the kernel; the hot component's fraction drops rapidly with increasing H g . Because of this, the apparent population fractions depend on the depth and ecliptic latitude of a transneptunian survey. The stirred and kernel components are limited to only a portion of the main belt, while we find that the hot component is consistent with a smooth extension throughout the inner, main and outer regions of the classical belt; in fact, the inner and outer belts are consistent with containing only hot-component objects. The H g ≤ 8.0 TNO population estimates are 400 for the inner belt and 10,000 for the outer belt to within a factor of two (95% confidence). We show how the CFEPS Survey Simulator can be used to compare a cosmogonic model for the the orbital element distribution to the real Kuiper belt. Subject headings: Kuiper Belt, surveys; PACS 96.30.Xa 8 9 42 non-resonant, non-scattering part of the belt beyond the 2:1 MMR with Neptune. Section 6 gives an order 43 of magntitude estimate of the scattering disk's population. Section 7 demonstrates the use of our Survey 44 -4 -Simulator to compare the results of a cosmogonic model to the CFEPS detections. Finally in Section 8, we 45 present our conclusions and put our findings in perspective. 46 2. Observations and Initial reductions 47 The discovery component of the CFEPS project imaged ∼320 square degrees of sky, almost all of 48 which was within a few degrees of the ecliptic plane. Discovery observations occurred in blocks of ≈ 16 49 fields acquired using the Canada-France-Hawaii Telescope (CFHT) MegaPrime camera which delivered ...