Lyman α emission from high‐redshift galaxies may be a powerful probe of the ionization history of the intergalactic medium (IGM) at z > 6; the observed Lyman α emission line is sensitive to the IGM hydrogen neutral fraction over the range 0.1–1. We present calculations of observed Lyman α emission lines from z > 6 galaxies, illustrating the effect of varying the many free parameters associated with the emitting galaxy, its halo and the IGM around the galaxy. In particular, we use a dynamic model of the IGM that includes the effect of IGM infall toward the emitting galaxy. Galactic winds may play a crucial role in determining observed Lyman α line fluxes. We compare our model predictions with observations of two z= 6.5 galaxies and conclude that, if galactic winds are allowed for, existing observations place no constraint on the neutral fraction of the IGM at z= 6.5. Future infrared observations will constrain the importance of galactic winds; if winds are unimportant for the observed z= 6.5 galaxies, our models suggest that the IGM neutral fraction at z= 6.5 is ≲0.1.

We calculate the contribution to the cosmic infrared background from very massive metal‐free stars at high redshift. We explore two plausible star formation models and two limiting cases for the reprocessing of the ionizing stellar emission. We find that Population III stars may contribute significantly to the cosmic near‐infrared background if the following conditions are met. (i) The first stars were massive, with M≳ 100 M⊙. (ii) Molecular hydrogen can cool baryons in low‐mass haloes. (iii) Population III star formation is ongoing, and not shut off through negative feedback effects. (iv) Virialized haloes form stars at ∼40 per cent efficiency up to the redshift of reionization, z∼ 7. (v) The escape fraction of the ionizing radiation into the intergalactic medium is small. (vi) Nearly all of the stars end up in massive black holes without contributing to the metal enrichment of the Universe.

We discuss the observational properties of a remarkably faint triply-imaged galaxy revealed in a deep z ′ -band Advanced Camera for Surveys observation of the lensing cluster Abell 2218 (z =0.175). A well-constrained mass model for the cluster, which incorporates the outcome of recent Keck spectroscopic campaigns, suggests that the triple system arises via a high redshift (z > 6) source viewed at high magnification (≃ ×25). Optical and infrared photometry from Hubble Space Telescope and the Keck Observatory confirms the lensing hypothesis and suggests a significant discontinuity occurs in the spectral energy distribution within the wavelength interval 9250-9850Å. If this break is associated with Gunn-Peterson absorption from neutral hydrogen, a redshift of 6.6 < z < 7.1 is inferred. Deep Keck spectroscopy conducted using both optical and infrared spectrographs fails to reveal any prominent emission lines in this region. However, an infrared stellar continuum is detected whose decline below 9800Å suggests a spectroscopic redshift towards the upper end of the range constrained photometrically, i.e. z ≃7. Regardless of the precise redshift, the source is remarkably compact ( < ∼ 1 h −1 70 kpc) and faint (z F 850LP =28.0) yet is undergoing vigorous star formation at a rate ≃2.6 M ⊙ yr −1 . An intriguing property is the steep slope of the ultraviolet continuum implied by the photometry which may suggest that the source is representative of an early population of galaxies responsible for cosmic reionization. Independent verification of these results is highly desirable but our attempts highlight the difficulty of studying such sources with present facilities and the challenges faced in pushing back the frontiers of the observable universe beyond z ∼6.5. 0 Using data obtained with the Hubble Space Telescope operated by AURA for NASA and the W.M. Keck Observatory on Mauna Kea, Hawaii. The W.M. Keck Observatory is operated as a scientific partnership among the California Institute of Technology, the University of California and NASA and was made possible by the generous financial support of the W.M. Keck Foundation.

We discuss the physical nature of a remarkably faint pair of Lyman α-emitting images discovered close to the giant cD galaxy in the lensing cluster Abell 2218 (z=0.18) during a systematic survey for highly-magnified star-forming galaxies beyond z=5. A well-constrained mass model suggests the pair arises via a gravitationally-lensed source viewed at high magnification. Keck spectroscopy confirms the lensing hypothesis and implies the unlensed source is a very faint (I ∼30) compact (<150 h −1 65 pc) and isolated object at z=5.576 whose optical emission is substantially contained within the Lyman α emission line; no stellar continuum is detectable. The available data suggest the source is a promising candidate for an isolated ∼10 6 M ⊙ system seen producing its first generation of stars close to the epoch of reionization. 1